1: \documentclass[12pt]{article}

2: \usepackage{amssymb}

3: \usepackage{graphics}

4: \usepackage{epsfig}

5: 

6: \def\baselinestretch{1.2}

7: \parskip 6 pt

8: 

9: \textheight 22.3 cm

10: %%\topmargin -.5 cm

11: %\topmargin -.8 cm

12: \topmargin -1.4cm

13: \textwidth 16cm

14: \oddsidemargin 0 in

15: \evensidemargin 0 in

16: 

17: 

18: \DeclareFontFamily{U}{rsf}{}

19: \DeclareFontShape{U}{rsf}{m}{n}{

20:   <5> <6> rsfs5 <7> <8> <9> rsfs7 <10-> rsfs10}{}

21: \DeclareMathAlphabet\Scr{U}{rsf}{m}{n}

22: 

23: %

24: %

25: % Putting citations together

26: %

27: \catcode`\@=11

28: %------------

29: %NAME: cite.sty

30: %

31: %              Replacement for \@citex

32: %

33: %     Allow, but strongly discourage, line breaks within a long

34: %     series of citations.  Compress lists of successive numbers

35: %     to one number range, e.g., 5,6,7,8,9 --> 5--9.  Compatible

36: %     with versions of \@cite that use exponents.

37: %              -- Donald Arseneau  1989

38: %

39: \def\@citex[#1]#2{%

40: \if@filesw \immediate \write \@auxout {\string \citation {#2}}\fi

41: \@tempcntb\m@ne \let\@h@ld\relax \def\@citea{}%

42: \@cite{%

43:   \@for \@citeb:=#2\do {%

44:     \@ifundefined {b@\@citeb}%

45:       {\@h@ld\@citea\@tempcntb\m@ne{\bf ?}%

46:       \@warning {Citation `\@citeb ' on page \thepage \space undefined}}%

47: %   else

48:       {\@tempcnta\@tempcntb \advance\@tempcnta\@ne%

49:       \@tempcntb\number\csname b@\@citeb \endcsname \relax%

50:       \ifnum\@tempcnta=\@tempcntb %Number follows previous--hold on to it

51:         \ifx\@h@ld\relax%

52: %         % first pair of successive

53:           \edef \@h@ld{\@citea\csname b@\@citeb\endcsname}%

54:         \else%

55: %         % compressible list of successive

56:           \edef\@h@ld{\ifmmode{-}\else--\fi\csname b@\@citeb\endcsname}%

57:         \fi%

58:       \else%   %  non-successor--dump what's held and do this one

59:         \@h@ld\@citea\csname b@\@citeb \endcsname%

60:         \let\@h@ld\relax%

61:       \fi}%

62:     \def\@citea{,\penalty\@highpenalty\,}%

63:   }\@h@ld

64: }{#1}}

65: 

66: %    \@cite macro using []'s:

67: %

68: \def\@citeb#1#2{{[#1]\if@tempswa , #2\fi}}

69: %

70: %    \@cite macro using exponents:

71: %

72: \def\@citeu#1#2{{$^{#1}$\if@tempswa , #2\fi }}

73: %

74: %    \@cite macro with nothing:

75: %

76: \def\@citep#1#2{{#1\if@tempswa , #2\fi}}

77: 

78: %

79: %   Style of citations:

80: %

81: \def\bcites{         % cite with []'s

82:         \catcode`\@=11

83:         \let\@cite=\@citeb

84:         \catcode`\@=12

85: }

86: 

87: \def\upcites{         % cite with exponents

88:         \catcode`\@=11

89:         \let\@cite=\@citeu

90:         \catcode`\@=12

91: }

92: 

93: \def\plaincites{      % cite without brackets

94:         \catcode`\@=11

95:         \let\@cite=\@citep

96:         \catcode`\@=12

97: }

98: 

99: %

100: %   Draft stuff

101: %

102: 

103: \newcount\hour

104: \newcount\minute

105: \newtoks\amorpm

106: \hour=\time\divide\hour by 60

107: \minute=\time{\multiply\hour by 60 \global\advance\minute by-\hour}

108: \edef\standardtime{{\ifnum\hour<12 \global\amorpm={am}%

109:         \else\global\amorpm={pm}\advance\hour by-12 \fi

110:         \ifnum\hour=0 \hour=12 \fi

111:         \number\hour:\ifnum\minute<10 0\fi\number\minute\the\amorpm}}

112: \edef\militarytime{\number\hour:\ifnum\minute<10 0\fi\number\minute}

113: 

114: \def\draftlabel#1{{\@bsphack\if@filesw {\let\thepage\relax

115:    \xdef\@gtempa{\write\@auxout{\string

116:       \newlabel{#1}{{\@currentlabel}{\thepage}}}}}\@gtempa

117:    \if@nobreak \ifvmode\nobreak\fi\fi\fi\@esphack}

118:         \gdef\@eqnlabel{#1}}

119: \def\@eqnlabel{}

120: \def\@vacuum{}

121: \def\marginnote#1{}

122: \def\draftmarginnote#1{\marginpar{\raggedright\scriptsize\tt#1}}

123: \overfullrule=0pt

124: 

125: \def\draft{

126:         \pagestyle{plain}

127:         \overfullrule=2pt

128:         \oddsidemargin -.5truein

129:         \def\@oddhead{\sl \phantom{\today\quad\militarytime} \hfil

130:         \smash{\Large\sl DRAFT} \hfil \today\quad\militarytime}

131:         \let\@evenhead\@oddhead

132:         \let\label=\draftlabel

133:         \let\marginnote=\draftmarginnote

134:         \def\ps@empty{\let\@mkboth\@gobbletwo

135:         \def\@oddfoot{\hfil \smash{\Large\sl DRAFT} \hfil}

136:         \let\@evenfoot\@oddhead}

137:         \def\@eqnnum{(\theequation)\rlap{\kern\marginparsep\tt\@eqnlabel}%

138:         \global\let\@eqnlabel\@vacuum}  }

139: 

140: 

141: 

142: %

143: %    Personal choice here

144: %

145: \def\section{\@startsection {section}{1}{\z@}{3.ex plus 1ex minus

146:  .2ex}{2.ex plus .2ex}{\large\bf}}

147: \def\subsection{\@startsection{subsection}{2}{\z@}{2.75ex plus 1ex minus

148:  .2ex}{1.5ex plus .2ex}{\bf}}

149: \def\subsect#1{\par\penalty1000{\noindent \bf #1}\par\penalty500}

150: \def\appendix{{\newpage\section*{Appendix}}\let\appendix\section%

151:         {\setcounter{section}{0}

152:         \gdef\thesection{\Alph{section}}}\section}

153: \def\thefootnote{\arabic{footnote}}

154: \def\abstract{\if@twocolumn

155: \section*{Abstract}

156: \else %\small

157: \begin{center}

158: {\bf Abstract\vspace{-.5em}\vspace{0pt}}

159: \end{center}

160: \quotation

161: \fi}

162: 

163: \catcode`\@=12

164: 

165: 

166: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

167: \newcommand{\beq}{\begin{equation}}

168: \newcommand{\eeq}{\end{equation}}

169: \newcommand{\beqa}{\begin{eqnarray}}

170: \newcommand{\eeqa}{\end{eqnarray}}

171: \newcommand{\dd}{{\rm d}}

172: \newcommand{\N}{{\bf N}}

173: \newcommand{\NN}{{\mathbb N}}

174: \newcommand{\Z}{{\bf Z}}

175: \newcommand{\ZZ}{{\mathbb Z}}

176: \newcommand{\Q}{{\bf Q}}

177: \newcommand{\QQ}{{\mathbb Q}}

178: \newcommand{\R}{{\bf R}}

179: \newcommand{\RR}{{\mathbb R}}

180: \newcommand{\C}{{\bf C}}

181: \newcommand{\CC}{{\mathbb C}}

182: \newcommand{\PP}{{\mathbb P}}

183: \newcommand{\e}{\,{\rm e}}

184: \newcommand{\CP}{{\CC\PP}}

185: \newcommand{\bra}{\langle}

186: \newcommand{\ket}{\rangle}

187: \newcommand{\no}{\nonumber}

188: \newcommand{\hsp}{\hskip5mm}

189: \newcommand{\vsp}{\vskip10mm}

190: \newcommand{\ds}{\displaystyle}

191: \newcommand{\ts}{\textstyle}

192: 

193: %  XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

194: %

195: %                Equation Definitions

196: %

197: \newcommand{\be}{\begin{eqnarray}}

198: \newcommand{\ee}{\end{eqnarray}}

199: \newcommand{\nn}{\nonumber}

200: 

201: %

202: %                 Math Relations

203: %

204: \def\da{\downarrow}

205: \def\ua{\uparrow}

206: \def\upda{\updownarrow}

207: \def\to{\rightarrow}

208: \def\implies{\Rightarrow}

209: \def\To{\longrightarrow}

210: \def\longlongrightarrow{\relbar\joinrel\relbar\joinrel\rightarrow}

211: \def\ridiculousrightarrow{\relbar\joinrel\relbar\joinrel\relbar%

212: \joinrel\rightarrow}

213: 

214: \def\underarrow#1{\mathrel{\mathop{\longrightarrow}\limits_{#1}}}

215: \def\onnearrow#1{\mathrel{\mathop{\nearrow}\limits^{#1}}}

216: \def\undernearrow#1{\mathrel{\mathop{\nearrow}\limits_{#1}}}

217: \def\onarrow#1{\mathrel{\mathop{\longrightarrow}\limits^{#1}}}

218: \def\onArrow#1{\mathrel{\mathop{\longlongrightarrow}\limits^{#1}}}

219: \def\OnArrow#1{\mathrel{\mathop{\ridiculousrightarrow}\limits^{#1}}}

220: \def\lae{\mathrel{\mathop{\smash{\lower .5 ex \hbox{$\stackrel<\sim$}}}}}

221: \def\lae{\mathrel{\mathop{\smash{\lower .5 ex \hbox{$\stackrel>\sim$}}}}}

222: \def\eqq{\stackrel?=}

223: 

224: %                        Math Stuff

225: 

226: \def\ket#1{\left| #1 \right\rangle}

227: \def\bra#1{\left\langle #1 \right|}

228: \def\vev#1{\left\langle #1 \right\rangle}

229: \def\VEV#1{\left\langle #1 \right\rangle}

230: \def\f{\frac}

231: \def\pa{\partial}

232: \def\pb{\bar\pa}

233: \def\na{\nabla}

234: \def\Tr{{\rm Tr}}

235: \def\tr{{\rm tr}}

236: \def\l:{\mathopen{:}\,}

237: \def\r:{\,\mathclose{:}}

238: \def\sech{\mathop{\rm sech}\nolimits}

239: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

240: 

241: %               DEFAULT SECTION SETUP

242: 

243: \catcode`\@=11

244: \def\theequation{\arabic{equation}}

245: %

246: %    for sections, subsections take from the following

247: %

248: %\def\theequation{\thesection.\arabic{equation}}

249: %\@addtoreset{equation}{section}

250: %\@addtoreset{footnote}{section}

251: %\@addtoreset{footnote}{subsection}

252: \catcode`\@=12

253: 

254: %                DEFAULT SETUP

255: 

256: 

257: \bcites

258: 

259: 

260: %                SETUP

261: 

262: %\draft

263: 

264: 

265: %             SECTION SETUP

266: 

267: \catcode`\@=11

268: \def\theequation{\thesection.\arabic{equation}}

269: \@addtoreset{equation}{section}

270: \@addtoreset{footnote}{section}

271: \@addtoreset{footnote}{subsection}

272: \catcode`\@=12

273: 

274: 

275: %               INFORMATION

276: 

277: \typeout{} \typeout{} \typeout{ } \typeout{} \typeout{}

278: \typeout{Why Kinks Look Like Monopoles.2} \typeout{} \typeout{}

279: \typeout{}

280: 

281: %

282: 

283: 

284: 

285: 

286: %

287: 

288: 

289: %

290: %           PAPER STARTS HERE

291: %

292: %load the bulking definitions etc.

293: %\input {shell}

294: 

295: 

296: \newcommand{\eps}{\epsilon}

297: \newcommand{\opsi}{\overline{\psi}}

298: \newcommand{\oQ}{\overline{Q}}

299: \newcommand{\btheta}{\overline{\theta}}

300: \newcommand{\cQ}{{\cal Q}}

301: \newcommand{\bepsilon}{\overline{\epsilon}}

302: \newcommand{\bPhi}{\overline{\Phi}}

303: \newcommand{\bP}{\overline{P}}

304: \newcommand{\bp}{\overline{p}}

305: \newcommand{\bphi}{\overline{\phi}}

306: \newcommand{\bpsi}{\overline{\psi}}

307: \newcommand{\bF}{\overline{F}}

308: \newcommand{\bJ}{\overline{J}}

309: \newcommand{\bh}{\overline{h}}

310: \newcommand{\bm}{\overline{m}}

311: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

312: \newcommand{\bi}{\overline{\imath}}

313: \newcommand{\bj}{\overline{\jmath}}

314: \newcommand{\bz}{\overline{z}}

315: \newcommand{\bchi}{\overline{\chi}}

316: \newcommand{\blambda}{\overline{\lambda}}

317: \newcommand{\bsigma}{\overline{\sigma}}

318: \newcommand{\bSigma}{\overline{\Sigma}}

319: \newcommand{\bareta}{\overline{\eta}}

320: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

321: \newcommand{\lrd}{\overleftarrow{\partial}\!\!\!\!\!

322: \overrightarrow{\vbox to 8.35pt{}}}

323: \newcommand{\lrD}{\overleftarrow{D}\!\!\!\!\!

324: \overrightarrow{\vbox to 8.2pt{}}}

325: \newcommand{\bD}{\overline{D}}

326: \newcommand{\bcQ}{\overline{\cal Q}}

327: \newcommand{\cD}{{\cal D}}

328: \newcommand{\rp}{{{\rm Re}\,p}}

329: \newcommand{\vp}{\varphi_P}

330: \newcommand{\cJ}{{\cal J}}

331: \newcommand{\rre}{{\rm Re}}

332: \newcommand{\iim}{{\rm Im}}

333: \newcommand{\ra}{\rightarrow}

334: \newcommand{\wh}{\widehat}

335: \newcommand{\bpartial}{{\bar\partial}}

336: \newcommand{\bA}{\overline{A}}

337: \newcommand{\tY}{{\tilde Y}}

338: \newcommand{\bY}{{\overline Y}}

339: \newcommand{\tL}{\tilde{L}}

340: \newcommand{\tG}{\tilde{G}}

341: \newcommand{\tJ}{\tilde{J}}

342: \newcommand{\tih}{\tilde{h}}

343: \newcommand{\tm}{\tilde{m}}

344: \newcommand{\tpsi}{\tilde{\psi}}

345: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

346: 

347: %%%%%%%%%%% New Commands %%%%%%%%%%%%%%%%

348: 

349: \newcommand\para{\paragraph{}}

350: \newcommand{\ft}[2]{{\textstyle\frac{#1}{#2}}}

351: \newcommand{\eqn}[1]{(\ref{#1})}

352: \def\Dslash{\,\,{\raise.15ex\hbox{/}\mkern-12mu D}}

353: \def\Dbarslash{\,\,{\raise.15ex\hbox{/}\mkern-12mu {\bar D}}}

354: \def\delslash{\,\,{\raise.15ex\hbox{/}\mkern-9mu \partial}}

355: \def\delbarslash{\,\,{\raise.15ex\hbox{/}\mkern-9mu {\bar\partial}}}

356: \def\pslash{\,\,{\raise.15ex\hbox{/}\mkern-9mu p}}

357: \def\calDslash{\,\,{\raise.15ex\hbox{/}\mkern-12mu {\cal D}}}

358: \newcommand{\sign}{{\rm sign}}

359: \newcommand{\vg}{\vec{g}}

360: \newcommand{\valpha}{\vec{\alpha}}

361: 

362: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

363: 

364: 

365: 

366: \begin{document}

367: \pagestyle{plain}

368: \setcounter{page}{1}

369: \newcounter{bean}

370: \baselineskip16pt

371: 

372: 

373: \begin{titlepage}

374: 

375: \begin{center}

376: 

377: 

378: \hfill\today\\

379: 

380: \vskip 3.0 cm {\Large On Monopoles and Domain Walls} \vskip 1 cm

381: {Amihay Hanany${}^1$ and David Tong${}^2$}\\

382: \vskip 1cm

383: {\sl ${}^1$ Center for Theoretical Physics, Massachusetts Institute of

384: Technology, \\

385: Cambridge, MA 02139, USA.\\}

386: {\tt hanany@mit.edu}\\

387: \vskip .5cm

388: {\sl ${}^2$ Department of Applied Mathematics and Theoretical Physics, \\

389: University of Cambridge, CB3 0WA, UK.} \\

390: {\tt d.tong@damtp.cam.ac.uk}

391: 

392: \end{center}

393: 

394: 

395: \vskip 1.5 cm

396: \begin{abstract}

397: The purpose of this paper is to describe a relationship between

398: maximally supersymmetric domain walls and magnetic monopoles. We

399: show that the moduli space of domain walls in non-abelian gauge

400: theories with $N$ flavors is isomorphic to a complex, middle

401: dimensional, submanifold of the moduli space of $U(N)$ magnetic

402: monopoles. This submanifold is defined by the fixed point set of a

403: circle action rotating the monopoles in the plane. To derive this

404: result we present a D-brane construction of domain walls, yielding

405: a description of their dynamics in terms of truncated Nahm

406: equations. The physical explanation for the relationship lies in

407: the fact that domain walls, in the guise of kinks on a vortex

408: string, correspond to magnetic monopoles confined by the Meissner

409: effect.

410: 

411: \end{abstract}

412: \end{titlepage}

413: 

414: 

415: \tableofcontents

416: 

417: \section{Introduction}

418: 

419: Domain walls in gauge theories with eight supercharges have rather

420: special properties. These walls were first studied by Abraham and

421: Townsend \cite{edpaul} who showed that in two-dimensions, where

422: domain walls are known as kinks, they exhibit dyonic behaviour

423: reminiscent of magnetic monopoles. Further similarities between

424: kinks and  magnetic monopoles, at both the classical and quantum

425: level, were uncovered in \cite{nick}. The physical explanation for

426: this relationship  was presented in \cite{stillme}, where new BPS

427: solutions were described corresponding to magnetic monopoles in a

428: phase with fully broken gauge symmetry. The Meissner effect

429: ensures that monopoles are confined: the magnetic flux is unable

430: to propagate through the vacuum and leaves the monopole in two

431: collimated tubes. From the perspective of the flux tube, the

432: monopole appears as a kink. The idea of describing confined

433: monopoles as kinks in $Z_N$ strings occurred previously in

434: \cite{bead}. The relationship between the confined magnetic

435: monopoles and the kink was further explored in

436: \cite{shif,meami,auzzi} and related systems were studied in

437: \cite{o1,o2,o3,o4,o5,o6}.

438: 

439: In this paper we use D-brane techniques to study the moduli space

440: of multiple domain walls. This allows us to develop a description

441: of the domain wall dynamics in terms of a linearized Nahm

442: equation, providing a direct relationship to the dynamics of

443: monopoles. Specifically, we show that the moduli space of domain

444: walls, which we denote as ${\cal W}_{\vg}$, is isomorphic to a

445: middle dimensional submanifold of the moduli space of magnetic

446: monopoles ${\cal M}_{\vg}$. This submanifold describes magnetic

447: monopoles lying along a line, and can be described as the fixed

448: point of an ${\bf S}^1$ action $\hat{k}$, rotating the monopoles

449: in a plane,

450: %

451: \be {\cal W}_{\vg}\cong\left.{\cal

452: M}_{\vg}\right|_{\hat{k}=0}\label{resu}\ee

453: %

454: The correspondence captures the topology and asymptotic metric of

455: the domain wall moduli space ${\cal W}_{\vg}$. It does not extend

456: to the full metric on ${\cal W}_{\vg}$. Nevertheless, as we shall

457: explain, it does correctly capture the most important feature of

458: domain walls: their ordering along the line.

459: 

460: The relationship \eqn{resu} plays companion to the result of

461: \cite{vib}, where the moduli space of vortices was shown to be a

462: middle dimensional submanifold of the moduli space of instantons.

463: Indeed, upon dimensional reduction,  the self-dual instanton

464: equations become the monopole equations, while the vortex

465: equations descend to the domain wall equations.

466: 

467: We start in the following section by describing the domain walls

468: in question, together with a review of their moduli spaces. We pay

469: particular attention to the crudest physical feature of domain

470: walls, namely the rules governing their spatial ordering along the

471: line. Section 3  contains a brief review of magnetic monopoles in

472: higher rank gauge groups, primarily in order to fix notation, allowing us

473: to elaborate on the relationship \eqn{resu}. We also describe the

474: Nahm construction of the monopole moduli space as it arises from

475: D-branes. The meat of the paper is in Section 4. We present a

476: D-brane embedding of domain wall solitons which gives a

477: description of their dynamics in terms of a linear Nahm equation.

478: This equation is somewhat trivial, with the content hidden in

479: various boundary conditions. We show how these boundary conditions

480: capture the prescribed ordering of domain walls.

481: 

482: 

483: 

484: \section{Domain Walls}

485: 

486: In this paper we will study a class of BPS domain wall solutions occurring in

487: maximally supersymmetric theories with multiple, isolated vacua. The Lagrangian

488: for these models includes a $U(k)$ gauge field $A_\mu$, a real adjoint

489: scalar $\sigma$ and $N$ fundamental scalars $q_i$, each with real mass $m_i$

490: %

491: \be {\cal L} ={\Tr} \left[\frac{1}{4e^2} F_{\mu\nu}F^{\mu\nu}

492: +\frac{1}{2e^2}|{\cal D}_\mu\sigma|^2 +\frac{e^2}{2}(q_i\!\otimes

493: \!q_i^\dagger-v^2)^2\right] +\sum_{i=1}^{N}\left[|{\cal D}_\mu

494: q_i|^2 +q_i^\dagger(\sigma-m_i)^2q_i \right]\nn\ee

495: %

496: where there is an implicit sum over the flavor index $i$ in the

497: adjoint valued term $q_i\otimes q_i^\dagger$. This Lagrangian can

498: be embedded in a theory with 8 supercharges in any spacetime

499: dimension $1\leq d\leq 5$ (e.g. ${\cal N}=2$ SQCD in $d=3+1$).

500: Such theories include further scalar fields which can be shown to

501: vanish on the domain wall solutions\footnote{If we promote the scalar field

502: $\sigma$ and the masses $m_i$ to complex variables, then the theories admit an

503: interesting array of domain wall junctions \cite{web} and dyonic walls \cite{14}.}.

504: The fermions do contribute zero modes but will not be important here.

505: 

506: When the Higgs expectation value $v^2$ is non-vanishing, and the

507: masses $m_i$ are distinct ($m_i\neq m_j$ for $i\neq j$), the

508: theory has a set of isolated vacua. Each vacuum is labelled by a

509: set $\Xi$ of $k$ distinct elements, chosen from a possible $N$,

510: %

511: \be \Xi = \left\{\xi(a): \xi(a)\neq \xi(b)\ {\rm for}\

512: a\neq b\right\} \label{set}\ee

513: %

514: Here $a=1,\ldots,k$ runs over the color index, while $\xi(a)\in

515: \{1,\ldots,N\}$. Up to a gauge transformation, the vacuum

516: associated to this set is given by,

517: %

518: \be \sigma={\rm diag}(m_{\xi(1)},\ldots,m_{\xi(k)}) \ \ \ \

519: \ \ ,\ \ \ \ \ \ \ q^a_{\ i}=v\,\delta^a_{\ i=\xi(a)}

520: \label{vacuum}\ee

521: %

522: For $N<k$ there are no supersymmetric vacua; for $N\geq

523: k$, the number of vacua is

524: %

525: \be N_{\rm vac}={ N\choose k}=\frac{N!}{k!(N-k)!} \label{nvac}\ee

526: %

527: Each of these vacua is isolated and exhibits a mass gap. There

528: are $k^2$ non-BPS massive gauge bosons and quarks with masses

529: $m^2_\gamma\sim e^2v^2+|m_{\xi(a)}-m_{\xi(b)}|^2$ and

530: $k(N-k)$ BPS massive quark fields with masses

531: $m_q\sim|m_{\xi(a)}-m_i|$ (with $i\notin \Xi$).

532: 

533: For vanishing masses $m_i=0$ the theory enjoys an $SU(N)$ flavor

534: symmetry, rotating the $q_i$. When distinct masses are turned on

535: this is broken explicitly to the Cartan-sub-algebra $U(1)^{N-1}$.

536: Meanwhile, the $U(k)$ gauge group is broken spontaneously in the

537: vacuum by the expectation values \eqn{vacuum}.

538: 

539: The existence of multiple, gapped, isolated vacua is sufficient to

540: guarantee the existence of co-dimension one domain walls

541: (otherwise known as kinks). These walls are BPS objects,

542: satisfying first order Bogomoln'yi equations which can be derived

543: in the usual manner by completing the square. We first choose a

544: flat connection $F_{\mu\nu}=0$, and allow the fields to depend

545: only on a single coordinate, say $x^3$. Then the Hamiltonian is

546: given by

547: %

548: \be {\cal H}&=& {\Tr} \left[\frac{1}{2e^2}|{\cal D}_3\sigma|^2 +

549: \frac{e^2}{2}(q_i\!\otimes \!q_i^\dagger-v^2)^2\right] +

550: \sum_{i=1}^{N}\left[|{\cal D}_3 q_i|^2

551: +q_i^\dagger(\sigma-m_i)^2q_i \right]\nn

552: \\  &=& \frac{1}{2e^2}\Tr\left[{\cal D}_3\sigma - e^2

553: (q_i\!\otimes\! q_i^\dagger -v^2)\right]^2 + \sum_{i=1}^{N} |{\cal

554: D}_3q_i-(\sigma-m_i)q_i|^2 \nn\\ && +\ \Tr \left[ {\cal

555: D}_3\sigma\,(q_i\!\otimes\!q_i^\dagger-v^2) \right]

556: +\sum_{i=1}^{N} \left[ q_i^\dagger(\sigma-m_i){\cal D}_3q_i+{\cal

557: D}_3q_i^\dagger(\sigma-m_i)q_i\right]

558: \\\nn &\geq& -\partial_3\left(v^2\,\Tr\,\sigma\right)

559: \ee

560: %

561: Our domain wall interpolates between a vacuum $\Xi_-$ at

562: $x^3=-\infty$, as determined by a set \eqn{set}, and a distinct

563: vacuum ${\Xi}_+$ at $x^3=+\infty$. The minus signs above have been

564: chosen under the assumption that $\Delta m>0$, where $\Delta

565: m=\sum_{i\in\,{\Xi}_-}m_i-\sum_{i\in\,\Xi_+}m_i$, so that a BPS

566: domain wall satisfies the Bogomoln'yi equations,

567: %

568: \be {\cal D}_3\sigma=e^2(q_i\!\otimes\!q_i^\dagger -v^2) \ \ \ \ \

569: , \ \ \ \ \ {\cal D}_3q_i=(\sigma-m_i)q_i \label{bog}\ee

570: %

571: and has tension given by $T=v^2\Delta m$. Analytic solutions to

572: these equations can be found in the $e^2\rightarrow \infty$ limit

573: \cite{me,isozumi,isonon}, which give smooth approximations to the

574: solution at large, but finite $e^2$ \cite{finite}.

575: 

576: 

577: \subsection{Classification of Domain Walls}

578: 

579: Domain walls in field theories are classified by the choice of

580: vacuum $\Xi_-$ and ${\Xi}_+$ at left and right infinity. However,

581: our theory contains an exponentially large number of vacua

582: \eqn{nvac} and one may hope that there is a coarser, less

583: unwieldy, classification which captures certain relevant

584: properties of a given domain wall. Such a classification was

585: offered in \cite{boojum}.

586: 

587: Firstly define the $N$-vector $\vec{m}=(m_1,\ldots, m_{N})$. The tension of the

588: BPS domain wall can then be written as

589: %

590: \be T_{\vg}=v^2\Delta m\equiv v^2\vec{m}\cdot \vec{g}

591: \label{tvg}\ee where the $N$-vector $\vec{g}\in\Lambda_R(su(N))$,

592: the root lattice of $su(N)$. Note that there do not exist domain

593: wall solutions for all $\vec{g}\in \Lambda_R(su(N))$; the only

594: admissible vectors are of the form $\vec{g}=(p_1,\ldots,p_{N})$

595: with $p_i=0$ or $\pm 1$. Note also that a choice of $\vec{g}$ does

596: not specify a unique choice of vacua $\Xi_-$ and ${\Xi}_+$ at left

597: and right infinity. Nor, in fact, does it specify a unique domain

598: wall moduli space ${\cal W}_{\vg}$. Nevertheless, domain walls in

599: sectors with the same $\vg$ share common traits.

600: 

601: The dimension of the moduli space of domain wall solutions was

602: computed in \cite{boojum} using an index theorem, following earlier

603: results in \cite{lee,isonon}.

604: To describe the dimension of the moduli space, it is useful to

605: decompose

606: $\vec{g}$ in terms of simple roots\footnote{The basis of simple

607: roots is fixed by the requirement that $\vec{m}\cdot

608: \vec{\alpha}_i >0$ for each $i$. A unique basis is defined in this

609: way if $\vec{m}$ lies in a positive Weyl chamber, which occurs

610: whenever the masses are distinct so that $SU(N)\rightarrow

611: U(1)^{N-1}$. If we choose the ordering $m_1>m_2>\ldots > m_{N}$ we

612: have simple roots $\vec{\alpha}_1=(1,-1,0,\ldots,0)$ and

613: $\vec{\alpha}_2=(0,1,-1,0,\ldots,0)$ through to

614: $\vec{\alpha}_{N-1}=(0,\ldots,1,-1)$.} $\vec{\alpha}_i$,

615: %

616: \be

617: \vec{g} =\sum_{i=1}^{N-1}n_i\vec{\alpha}_i\ \ \ \  \ \ , \ \ \ \ \ \ n_i\in\Z

618: \label{dwg}\ee

619: %

620: The index theorem of

621: \cite{boojum} reveals that domain wall solutions to \eqn{bog}

622: exist only if $n_i\geq 0$ for all $i$. If this holds, the number

623: of zero modes of a solution is given by

624: %

625: \be

626: {\rm dim}\left({\cal W}_{\vec{g}}\right) = 2\sum_{i=1}^{N-1}n_i

627: \label{dwdim}\ee

628: %

629: where ${\cal W}_{\vg}$ denotes the moduli space of any set of

630: domain walls with charge $\vg$. This result has a simple physical

631: interpretation. There exist $N-1$ types of ``elementary'' domain

632: walls corresponding to a $\vec{g}=\vec{\alpha}_i$, the simple

633: roots. Each of these has just two collective coordinates

634: corresponding to a position in the $x^3$ direction and a phase. As

635: first explained in \cite{edpaul}, the phase coordinate is a

636: Goldstone mode arising because the domain wall configuration

637: breaks the $U(1)^{N-1}$ flavor symmetry as we review below. In

638: general, a domain wall labelled by $\vec{g}$ can be thought to be

639: constructed from $\sum_in_i$ elementary domain walls, each with

640: its own position and phase collective coordinate. Let us now turn

641: to some examples.

642: 

643: \subsection{The Moduli Space of Domain Walls: Some Examples}

644: 

645: \subsubsection*{\it Example 1: $\vg=\valpha_1$}

646: 

647: The simplest system admitting a domain wall is the abelian $k=1$ theory with

648: $N=2$ charged scalars $q_1$ and $q_2$. The $N_{vac}=2$ vacua of the

649: theory are given by $\sigma=m_i$ and $|q_j|^2=v^2\delta_{ij}$ for $i=1,2$.

650: There is a single domain wall in this theory with $\vec{g}=\vec{\alpha}_1$ interpolating

651: between these two vacua. Under the $U(1)_F$ flavor symmetry, $q_1$ has charge $+1$ and

652: $q_2$ has charge $-1$. In each of the vacua, the $U(1)_F$ symmetry coincides

653: with the $U(1)$ gauge action but, in the core of the domain wall, both $q_1$ and

654: $q_2$ are non-vanishing, and $U(1)_F$ acts non-trivially. The resulting goldstone

655: mode is the phase collective coordinate.

656: The moduli space of the domain wall is simply

657: %

658: \be {\cal W}_{\vec{g}=\vec{\alpha}_1}\cong\R\times{\bf S}^1

659: \ee

660: %

661: where the $\R$ factor describes the center of mass of the domain wall and

662: the ${\bf S}^1$ the phase. One can show that the ${\bf S}^1$ has radius

663: $2\pi v^2/T_{\vg}=2\pi /(m_1-m_2)$.

664: 

665: \subsubsection*{\it Example 2: $\vg=\valpha_1+\valpha_2$}

666: 

667: The simplest system admitting multiple domain walls is the abelian

668: $k=1$ theory with $N=3$ charged scalars. There are now three

669: vacua, given by $\sigma=m_i$ and $|q_j|^2=v^2\delta_{ij}$. In each

670: a $U(1)_{F_1}\times U(1)_{F_2}$ flavor symmetry is unbroken, under

671: which the scalars have charge

672: %

673: \be U(1)_{F_1}\times U(1)_{F_2}: \left\{\begin{array}{cl} q_1:& (+1,0) \\ q_2: &

674: (-1,1) \\ q_3: & (0,-1) \end{array}\right.\ee

675: %

676: The first elementary domain wall $\vec{g}=\vec{\alpha}_1$ interpolates between vacuum 1

677: and vacuum 2, breaking $U(1)_{F_1}$ along the way. The second elementary domain wall

678: interpolates between vacuum 2 and vacuum 3, breaking $U(1)_{F_2}$ along the way. Of interest

679: here is the domain wall $\vec{g}=\vec{\alpha}_1+\vec{\alpha}_2$ interpolating between

680: vacuum 1 and vacuum 3. It can be thought of as a composite of two domain walls.

681: The moduli space for these two domain walls

682: was studied in \cite{me,mmotw} and is of the form,

683: %

684: \be

685: {\cal W}_{\vg=\valpha_1+\valpha_2}\cong {\bf R}\times \frac{\R \times

686: \tilde{\cal W}_{\valpha_1+\valpha_2}}{\Z}

687: \label{12}\ee

688: %

689: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

690: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

691: \newcommand{\onefigurenocap}[1]{\begin{figure}[h]

692:          \begin{center}\leavevmode\epsfbox{#1.eps}\end{center}

693:          \end{figure}}

694: \newcommand{\onefigure}[2]{\begin{figure}[htbp]

695:          \begin{center}\leavevmode\epsfbox{#1.eps}\end{center}

696:          \caption{\small #2\label{#1}}

697:          \end{figure}}

698: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

699: \begin{figure}[tbp]

700: \begin{center}

701: \epsfxsize=3.0in\leavevmode\epsfbox{mspace.eps}

702: \end{center}

703: \begin{center}

704: \begin{minipage}{13cm}

705:  \caption{\small %Figure 1:

706: The relative moduli space $\tilde{W}_{\valpha_1+\valpha_2}$ of two domain walls

707: is a cigar.}

708: \end{minipage}

709: \end{center}

710: \label{fig:reconnect}

711: \end{figure}

712: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

713: The first factor of $\R$ corresponds to the center of mass of the

714: two domain walls; the second factor corresponds to the combined

715: phase associated to the two domain walls. When the ratio of

716: tensions of the two elementary domain walls

717: $T_{\valpha_1}/T_{\valpha_2}$ is rational, the ratio of the

718: periods of the two phases are similarly rational and the second

719: $\R$ factor collapses to ${\bf S}^1$, while the quotient $\Z$

720: reduces to a finite group. The relative moduli space $\tilde{\cal

721: W}_{\valpha_1+\valpha_2}$ corresponds to the separation and

722: relative phases of the two domain walls. Importantly, and unlike

723: other solitons of higher co-dimension, the domain walls must obey

724: a strict ordering on the $x^3$ line: the $\vg=\valpha_1$ domain

725: wall must always be to the left of $\vg=\valpha_2$ domain wall.

726: The separation between the walls is therefore the halfline $\R^+$.

727: It was shown in \cite{me} that the relative phase is fibered over

728: $\R^+$ to give rise to a smooth cylinder, with the tip

729: corresponding to coincident domain walls. The resulting moduli

730: space is shown in Figure 1.

731: 

732: Note that the moduli space \eqn{12} is toric, inheriting two

733: isometries from the $U(1)_{F_1}\times U(1)_{F_2}$ symmetry. In an

734: abelian gauge theory with arbitrary number of flavors $N$, the

735: domain wall charge is always of the form $\vg=\sum_in_i\valpha_i$

736: with $n_i=0,1$, and the moduli space is always toric, meaning that

737: half of the dimensions correspond to $U(1)$ isometries.

738: 

739: 

740: 

741: \subsubsection*{\it Example 3: $\vg=\valpha_1+2\valpha_2+\valpha_3$}

742: 

743: In non-abelian theories, the domain wall moduli spaces are no

744: longer toric. The simplest such theory has a $U(2)$ gauge group

745: with $N=4$ fundamental scalars. The 6 vacua, and 15 different

746: domain walls, of this theory were detailed in \cite{boojum}. Under

747: the $U(1)^3_F$ flavor symmetry, the fundamental scalars transform

748: as

749: %

750: \be

751: U(1)_{F_1}\times U(1)_{F_2}\times U(1)_{F_3}:\left\{\begin{array}{cl}

752: q_1:& (+1,0,0) \\ q_2: & (-1,1,0) \\ q_3: & (0,-1,1) \\ q_4: & (0,0,-1)\\

753: \end{array}\right.

754: \label{3u1s}\ee

755: %

756: With this convention, the elementary domain wall $\vg=\valpha_i$ picks up its

757: phase from the action of the $U(1)_{F_i}$ flavor symmetry.

758: 

759: Here we concentrate on the domain wall system with the maximal

760: number of zero modes which arises from the choice of vacua

761: ${\Xi}_-=(1,2)$ and $\Xi_+=(3,4)$ so that $\vg =

762: \valpha_1+2\valpha_2+\valpha_3$. This system can be separated into

763: four constituent domain walls. As explained in \cite{isonon,boojum},

764: the ordering of domain walls is no longer strictly

765: fixed in this example. The two outer elementary domain walls, on

766: the far left and far right, are each of the type $\vg=\valpha_2$.

767: However, the relative positions of the middle two domain walls,

768: $\vg=\valpha_1$ and $\valpha_3$ are not ordered and they may pass

769: through each other.

770: 

771: Unlike the situation for abelian gauge theories, the 8 dimensional

772: domain wall moduli space for this example is no longer toric;

773: ${\cal W}_{\vg}$ inherits only three $U(1)$ isometries from

774: \eqn{3u1s}. Physically this means that the two phases associated

775: to the $\valpha_2$ domain walls are not both Goldstone modes and

776: they may interact as the domain walls approach. This behaviour

777: is familiar from the study of the Atiyah-Hitchin metric describing the

778: dynamics of two monopoles in $SU(2)$ gauge theory; we shall make

779: the analogy more precise in the following.

780: 

781: 

782: \subsection{The Ordering of Domain Walls}

783: 

784: As we stressed in the above examples, in contrast to other

785: solitons domain walls must obey some ordering on the line. This

786: will be an important ingredient when we come to extract domain

787: wall data from the linearized Nahm's equations in Section 4. Here

788: we linger to review this ordering.

789: 

790: %Firstly, let us define the ``maximal domain wall'' which, for

791: %fixed $k$ and $N$, interpolates between $\Xi_-=\{1,2,\ldots, k\}$

792: %and ${\Xi}_+=\{N-k+1,\ldots, N\}$. For $N\geq 2k$ the maximal

793: %domain wall corresponds to the topological charge vector

794: %\be\vg=(1,1,\ldots,1,0,\ldots, 0,-1,\ldots,-1,-1)\ee where there

795: %are $k$ $1$'s and $k$ $(-1)$'s. In terms of the co-roots

796: %decomposition we have

797: %%

798: %\be

799: %n_i=\left\{\begin{array}{ll} i\ \ \ & i=1\ldots, k \\ k &

800: %i=k,\ldots, N-k \\ N-i & i=N-k,N-1\end{array}\right.

801: %\label{maximal}\ee

802: %%

803: %For $N<2k$ the $n_i$ increase and decrease in unit amounts, never

804: %reaching the plateau in the middle. In what follows we shall focus

805: %on this maximal sector. There is no loss of generality in this;

806: %all other sectors are either subsets of this sector, reached by

807: %removing some of the outer domain walls to infinity, or are

808: %non-interacting products of these subsets. Each of the examples of

809: %the previous subsection was a maximal domain wall.

810: 

811: The ordering of the elementary domain walls in non-abelian

812: theories was studied in detail in \cite{isonon}. One can derive

813: the result by considering the possible sequences of vacua as we

814: move over each domain wall. For example, we could consider the

815: ``maximal domain wall'', interpolating between

816: $\Xi_-=\{1,2,\ldots,k\}$ and $\Xi_+=\{N-k+1,\ldots,N\}$. From the

817: left, the first elementary domain wall that we come across must be

818: $\vec{g}=\valpha_{k}$, corresponding to

819: $\Xi_-=(1,2,\ldots,k-1,k)\rightarrow (1,2,\ldots,k-1,k+1)$. The

820: next elementary domain wall may be either $\valpha_{k-1}$ or

821: $\valpha_{k+1}$. These two walls are free to pass through each

822: other, but cannot move further to the left than the $\valpha_k$

823: wall. And so on. Iterating this procedure, one finds that two

824: neighbouring elementary domain walls $\valpha_i$ and $\valpha_j$

825: may pass through each other whenever $\valpha_i\cdot\valpha_j=0$,

826: but otherwise have a fixed ordering on the line. The net result of

827: this analysis is summarized in Figure 2. The $x^3$ positions of

828: the domain walls are shown on the vertical axis; the position on

829: the horizontal axis denotes the type of elementary domain wall,

830: starting on the left with $\valpha_1$ and ending on the right with

831: $\valpha_{N-1}$.

832: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

833: \begin{figure}[tbp]

834: \begin{center}

835: \epsfxsize=14cm\leavevmode\epsfbox{ordering.eps}

836: \end{center}

837: \begin{center}

838: \begin{minipage}{15cm}

839:  \caption{\small

840: The ordering for the maximal domain wall. The $x^3$ spatial

841: direction is shown horizontally. The position in the vertical

842: direction denotes the type of domain wall. Domain walls of

843: neighbouring types have their positions interlaced.}

844: \end{minipage}

845: \end{center}

846: \label{fig:ninety   }

847: \end{figure}

848: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

849: 

850: In summary, we see that for $n_i=n_{i+1}-1$, the $\valpha_i$

851: domain walls are trapped between the $\valpha_{i+1}$ domain walls.

852: The reverse holds when $n_i=n_{i+1}+1$. Finally, when

853: $n_{i}=n_{i+1}$ the positions of the $\valpha_{i}$ domain walls

854: are interlaced with those of the $\valpha_{i+1}$ domain walls. The

855: last $\valpha_{i+1}$ domain wall is unconstrained by the

856: $\valpha_i$ walls in its travel in the positive $x^3$ direction,

857: although it may be trapped in turn by a $\valpha_{i+2}$ wall.

858: 

859: 

860: While we have discussed the maximal domain wall above, other sectors can

861: be reached either by removing some of the outer domain walls to infinity,

862: or by taking non-interacting products such subsets. It's important to note

863: that labelling a topological sector $\vg$ does not necessarily determine

864: the ordering of domain walls\footnote{An example: the

865: $\vg=\valpha_1+\valpha_2+\valpha_3$ domain wall. In the abelian theory with

866: $k=1$ and $N=4$, the ordering is $\valpha_1<\valpha_2<\valpha_3$. However,

867: in the non-abelian theory with $k=2$ and $N=4$, the ordering is

868: $\valpha_1,\valpha_3<\valpha_2$.}. We shall show that domain walls with the

869: same $\vg$, but different orderings, descend from different submanifolds

870: of the same monopole moduli space.

871: 

872: \section{Monopoles}

873: 

874: The main goal of this paper is to show how the moduli space of

875: domain walls introduced in the previous section is isomorphic to a

876:  submanifold of a related monopole moduli space. In this section

877: we review several relevant aspects of these monopole moduli

878: spaces.

879: 

880: It will turn out that

881: the domain walls of Section 2 are related to monopoles in an

882: $SU(N)$ gauge theory. Note that the flavor group from Section 2

883: has been promoted to a gauge group; we shall see the reason behind

884: this in Section 5. The Bogomoln'yi monopole equations are

885: %

886: \be B_\mu={\cal D}_\mu\phi\label{monbog}\ee

887: %

888: where $B_\mu$, $\mu=1,2,3$ is the $SU(N)$ magnetic field and $\phi$ is an adjoint

889: valued real scalar field. The monopoles exist only if $\phi$ takes a

890: vacuum expectation value,

891: %

892: \be \vev{\phi} ={\rm diag} (m_1,\ldots, m_{N})

893: \label{mono}\ee

894: %

895: where we take $m_i\neq m_j$ for $i\neq j$, ensuring breaking to

896: the maximal torus, $SU(N)\rightarrow U(1)^{N-1}$. It is not

897: coincidence that we've denoted the vacuum expectation values by

898: $m_i$, the same notation used for the masses in Section 2; it is

899: for this choice of vacuum that the correspondence holds.

900: (Specifically, the masses of the kinks will coincide with the masses

901: of monopoles, ensuring that the asymptotic metrics on ${\cal W}_{\vg}$

902: and ${\cal M}_{\vg}$ also coincide).

903: 

904: 

905: As described long ago by Goddard, Nuyts and Olive \cite{gno}, the

906: allowed magnetic charges under each unbroken $U(1)^{N-1}$ are

907: specified by a root vector\footnote{We ignore the factor of 2

908: difference between roots and co-roots. For simply laced groups,

909: such as $SU(N)$, it can be absorbed into convention.} of $su(N)$,

910: $\vg=(p_1,\ldots,p_{N})$. It is customary to decompose this in

911: terms of simple roots $\valpha_i$,

912: %

913: \be

914: \vec{g} =\sum_{i=1}^{N-1}n_i\vec{\alpha}_i\ \ \ \  \ \ , \ \ \ \ \ \ n_i\in\Z

915: \label{mg}\ee

916: %

917: Once again, the notation is identical to that used for domain

918: walls \eqn{dwg} for good reason. Solutions to the monopole

919: equations \eqn{monbog} exist for all values of $n_i\geq 0$.

920: This is in contrast to domain walls where, as we have seen, configurations

921: only exist in a finite number of sectors defined by $p_i=0$ or

922: $p_i=\pm 1$.  The mass of the magnetic monopole is

923: $M_{\rm mono}=(2\pi/e^2)\vec{m}\cdot\vg$.

924: 

925: The monopole moduli space ${\cal M}_{\vg}$ is the space of solutions to \eqn{mono}

926: in a fixed topological sector $\vg$. The dimension of this space, equal to the number

927: of zero modes of given solution, was computed by E. Weinberg in \cite{erick} using

928: Callias' version of the index theorem. The result is:

929: %

930: \be {\rm dim}\,({\cal M}_{\vg})=4\sum_{i=1}^{N-1}n_i

931: \label{mondim}\ee

932: %

933: which is to be compared with \eqn{dwdim}.

934: 

935: \subsection{The Relationship between ${\cal M}_{\vg}$ and ${\cal W}_{\vg}$}

936: 

937: We are now in a position to describe the relationship between the moduli space of

938: domain walls ${\cal W}_{\vg}$ and the moduli space of magnetic monopoles ${\cal M}_{\vg}$.

939: We will show that ${\cal W}_{\vg}$ is a complex, middle dimensional, submanifold of

940: ${\cal M}_{\vg}$, defined by the fixed point set of the action rotating the monopoles

941: in a plane, together with a suitable gauge action.

942: To do this, we first need to describe the symmetries of ${\cal M}_{\vg}$.

943: 

944: The monopole moduli space ${\cal M}_{\vg}$ admits a natural,

945: smooth, hyperK\"ahler metric \cite{manton,ah}. For generic $\vg$

946: this metric enjoys $(N-1)$ tri-holomorphic isometries arising from

947: the action of the $U(1)^{N-1}$ abelian gauge group. Further the

948: metric has an $SU(2)_R$ symmetry, arising from rotations of the

949: monopoles in $\R^3$, which acts on the three complex structures of

950: ${\cal M}_{\vg}$. In other words, any $U(1)_R\subset SU(2)_R$ is a

951: holomorphic isometry, preserving a single complex structure while

952: revolving the remaining two. Let us choose $U(1)_R$ to rotate the

953: monopoles in the $(x^2-x^3)$ plane. In what follows we will be

954: interested in a specific holomorphic $\hat{U}(1)$ action which

955: acts simultaneously by a $U(1)_R$ rotation and a linear

956: combination of the gauge rotations $U(1)^{N-1}$ (to be specified

957: presently). We denote the Killing vector on ${\cal M}_{\vg}$

958: associated to $\hat{U}(1)$ as $\hat{k}$. We claim

959: %

960: \be {\cal W}_{\vg}=\left.{\cal M}_{\vg}\right|_{\hat{k}=0}

961: \label{result}\ee

962: %

963: This result holds at the level of topology and asymptotic metric

964: of the spaces. The manifold ${\cal W}_{\vg}$ inherits a metric

965: from ${\cal M}_{\vg}$ by this reduction: it does not coincide with

966: the domain wall metric in the interior on ${\cal W}_{\vg}$. (For

967: example, corrections to the asymptotic metric on ${\cal W}_{\vg}$ are

968: exponentially suppressed while those of ${\cal M}_{\vg}$ have

969: power law behaviour). It would be interesting to examine if ${\cal

970: W}_{\vg}$ inherits the correct K\"ahler class and/or complex

971: structure from ${\cal M}_{\vg}$.

972: 

973: We defer a derivation of \eqn{result} to the following section, but first present

974: some simple examples.

975: 

976: \subsubsection*{\it Example 1: $\vg=\valpha_1$}

977: 

978: Monopoles in $SU(2)$ gauge theories are labelled by a single topological

979: charge $\vg=n_1\valpha_1$. For a single monopole ($n_1=1$) the moduli

980: space is simply

981: %

982: \be

983: {\cal M}_{\vg=\valpha_1}\cong\R^3\times {\bf S}^1

984: \ee

985: %

986: where the $\R^3$ factor denotes the position of the monopole,

987: while the ${\bf S}^1$ arises from global gauge transformations

988: under the surviving $U(1)$. The radius of the ${\bf S}^1$ is

989: $2\pi /(m_1-m_2)$. In this case the $\hat{U}(1)$ action coincides with

990: the rotation $U(1)_R$ in the $(x^2-x^3)$ plane and we have

991: trivially

992: %

993: \be

994: {\cal W}_{\valpha_1}\cong \R\times {\bf S}^1\cong

995: \left.{\cal M}_{\valpha_1}\right|_{\hat{k}=0}

996: \ee

997: %

998: The similarity between the domain wall and monopole moduli spaces

999: for a single soliton was noted by Abraham and Townsend

1000: \cite{edpaul}. In both cases, motion in the ${\bf S}^1$ factor

1001: gives rise to dyonic solitons.

1002: 

1003: Note that monopole moduli spaces for charges $\vg=n_1\valpha_1$

1004: exist for all $n_1\in\Z^+$. For example, the $n_1=2$ monopole

1005: moduli space is home to the famous Atiyah-Hitchin metric

1006: \cite{ah}. However, there is no domain wall moduli space with this

1007: charge in the class of theories we discuss in Section 2.

1008: 

1009: \subsubsection*{\it Example 2: $\vg=\valpha_1+\valpha_2$}

1010: 

1011: Our second example is the $\vg=\valpha_1+\valpha_2$ monopole in

1012: $SU(3)$ gauge theories (sometimes referred to as the $(1,1)$ monopole). The

1013: moduli space was determined in \cite{conell,gl,lwy1} to be of the

1014: form

1015: %

1016: \be

1017: {\cal M}_{\vg=\valpha_1+\valpha_2}\cong\R^3\times\frac{\R\times

1018: \tilde{\cal M}_{\valpha_1+\valpha_2}}{\Z}

1019: \ee

1020: %

1021: where the relative moduli space $\tilde{\cal M}_{\valpha_1+\valpha_2}$ is the

1022: Euclidean Taub-NUT space, endowed with the metric

1023: %

1024: \be

1025: ds^2=\left(1+\frac{1}{r}\right)\,(dr^2+r^2d\theta^2+^2\sin^2\theta\,d\phi^2)

1026: +\left(1+\frac{1}{r}\right)^{-1}(d\psi+\cos\theta d\phi)^2

1027: \label{tn}\ee

1028: %

1029: Here $r$,$\theta$ and $\phi$ are the

1030: familiar spherical polar coordinates. The coordinate $\psi$ arises

1031: from $U(1)$ gauge transformations. The manifold has a $SU(2)_R\times U(1)$

1032: isometry, of which only a $U(1)_R\times U(1)$ are manifest in the above

1033: coordinates. The holomorphic $U(1)_R$ isometry acts by rotating the two

1034: monopoles: $\phi\rightarrow \phi+c$. The tri-holomorpic $U(1)$ isometry

1035: changes the relative phase of the monopoles: $\psi\rightarrow \psi+c$.

1036: Both of these actions have a unique fixed point at $r=0$, the ``nut'' of Taub-NUT.

1037: However, the combined action with Killing vector $\partial_\psi+\partial_\phi$

1038: has a fixed point along the half-line $\theta=\pi$, with $\psi$ fibered over this

1039: line to produce the cigar shown in Figure 1. This is the relative moduli space

1040: $\tilde{\cal W}_{\valpha_1+\valpha_2}$.

1041: 

1042: 

1043: Similar calculations hold for monopoles of charge

1044: $\vg=\sum_{i=1}^{N-1}\valpha_i$, whose dynamics is described by a

1045: class of toric hyperK\"ahler metrics, known as the Lee-Weinberg-Yi

1046: metrics \cite{lwy2}. Once again, a suitable ${\bf S}^1$ action on

1047: these spaces can be identified such that the fixed points localize

1048: on ${\cal W}_{\vg}$, the moduli space of domain walls in $U(1)$

1049: gauge theories with $N$ charged scalars.

1050: 

1051: 

1052: 

1053: \subsubsection*{\it Example 3: $\vg=\valpha_1+2\valpha_2+\valpha_1$}

1054: 

1055: As described in the previous section, the simplest domain wall charge

1056: $\vg=\sum_in_i\valpha_i$ with some $n_i>1$ occurs for $\vg=\valpha_1+

1057: 2\valpha_2+\valpha_3$, and corresponds to a monopole in a $SU(4)$

1058: gauge theory. No explicit expression for the  metric on this monopole moduli space is

1059: known although, given the results of \cite{21}, such a computation may be

1060: feasible. Without an explicit expression for the metric in this, and more

1061: complicated examples, we need a more powerful method to describe the

1062: moduli space. This is provided by the Nahm construction,

1063: which we now review

1064: 

1065: \subsection{D-Branes and Nahm's Equations}

1066: 

1067: The moduli space of magnetic monopoles is isomorphic to the moduli

1068: space of Nahm data. Here we review the Nahm construction of the

1069: monopole moduli space \cite{nahm} and, in particular, the

1070: embedding within the framework of D-branes due to Diaconescu

1071: \cite{diac}. This will be useful to compare to the domain walls of

1072: the next section.

1073: 

1074: In the D-brane construction, Nahm's equations arise as the

1075: low-energy description of D-strings suspended between D3-branes

1076: \cite{diac}. The $SU(N)$ Yang-Mills theory lives on the

1077: worldvolume of $N$ D3-branes separated in, say, the $x_6$

1078: direction, with the $i^{\rm th}$ D3-brane placed at position

1079: $(x_6)_i=m_i$ in accord with the adjoint expectation value

1080: \eqn{mono}. The monopole of charge $\vg=\sum_in_i\valpha_i$

1081: corresponds to suspending $n_i$ D-strings between the $i^{\rm th}$

1082: and $(i+1)^{\rm th}$ D3-brane. This configuration is shown in

1083: Figure 3.

1084: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1085: \begin{figure}[tbp]

1086: \begin{center}

1087: \epsfxsize=10cm\leavevmode\epsfbox{monopole.eps}

1088: \end{center}

1089: \begin{center}

1090: \begin{minipage}{15cm}

1091:  \caption{\small

1092: The $\vg=3\valpha_1+2\valpha_2+3\valpha_3$ monopole as D-strings

1093: stretched between D3-branes.}

1094: \end{minipage}

1095: \end{center}

1096: \end{figure}

1097: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1098: 

1099: The motion of the D-strings in each segment $m_i\leq x_6\leq m_{i+1}$ is

1100: governed by four hermitian $n_i\times n_i$  matrices,

1101: $X_1,X_2,X_3$ and $A_6$ subject to the covariant version of Nahm's equations,

1102: \be

1103: \frac{dX_\mu}{dx_6}-i[A_6,X_\mu]-\frac{i}{2}\epsilon_{\mu\nu\rho}

1104: [X_\nu,X_\rho]=0\ \ \ \ \ \ \ \ m_i\leq x_6\leq m_{i+1}

1105: \label{nahm}\ee

1106: %

1107: %\be

1108: %\frac{dX^{[i]}_\mu}{dx_6}-i[X^{[i]}_0,X^{[i]}_\mu]-\fti2\epsilon_{\mu\nu\rho}

1109: %[X^{[i]}_\nu,X^{[i]}_\rho]=0

1110: %\ee

1111: %%

1112: modulo $U(n_i)$ gauge transformations acting on the interval $m_i\leq

1113: x_6\leq m_{i+1}$, and vanishing  at the boundaries. The $X_\mu$ form the

1114: triplet representation of the $SU(2)_R$ symmetry which rotates monopoles in $\R^3$.

1115: The $U(1)^{N-1}$ surviving gauge transformations acts on the Nahm data by constant

1116: shifts of the $(N-1)$ ``Wilson lines'' $A_6\rightarrow A_6 + c1_{n_i}$.

1117: 

1118: The interactions

1119: between neighbouring segments depends on the relative size of

1120: the matrices and is given by \cite{hub}

1121: 

1122: \underline{\ $n_i=n_{i+1}$:\ } In this case the $U(n_i)$ gauge symmetry is

1123: extended to the interval $m_i\leq x_6\leq m_{i+2}$ and an impurity is

1124: added to the right-hand-side of Nahm's equations, which now read

1125: %

1126: \be

1127: \frac{dX_\mu}{dx_6}-i[A_6,X_\mu]-\frac{i}{2}\epsilon_{\mu\nu\rho}

1128: [X_\nu,X_\rho]=\omega_\alpha\, \sigma_\mu^{\alpha\beta}\,\omega^\dagger_\beta\

1129: \delta(x_6-m_{i+1})

1130: \label{imp}\ee

1131: %

1132: Here $\sigma_\mu$ are the Pauli matrices. The impurity degrees of freedom

1133: lie in the complex 2-vector, $\omega_\alpha=(\psi,\tilde{\psi}^\dagger)$ which

1134: is a doublet under the $SU(2)_R$ symmetry. Both $\psi$ and $\tilde{\psi}^\dagger$

1135: are themselves complex $n_i$

1136: vectors, transforming in the fundamental representation of the $U(n_i)$

1137: gauge group. The combination

1138: $\omega_\alpha \sigma_\mu^{\alpha\beta}\omega^\dagger_\beta$ is thus

1139: an $n_i\times n_i$

1140: matrix, transforming in the adjoint representation of the gauge group.

1141: The $\omega_\alpha$ fields can be thought of as a hypermultiplet

1142: arising from $D1-D3$ strings \cite{hw,kapset,tsimpis}

1143: 

1144: \underline{\ $n_i=n_{i+1}-1$:\ } In this case $X_\mu\rightarrow (X_\mu)_-$, a set

1145: of three $n_i\times n_i$ matrices, as $x_6\rightarrow (m_i)_-$ from the left. To

1146: the right of $m_i$, the $X_\mu$ are $(n_i+1)\times (n_i+1)$ matrices which

1147: must obey

1148: %

1149: \be

1150: X_\mu\rightarrow \left(\begin{array}{cc} y_\mu & a_\mu^\dagger \\ a_\mu & (X_\mu)_-

1151: \end{array}\right)\ \ \ \ \ {\rm as\ } x_6\rightarrow (m_i)_+

1152: \ee

1153: %

1154: where $y_\mu\in\R$ and each $a_\mu$ is a complex $n_i$-vector.

1155: 

1156: \underline{\ $n_i\leq n_{i+1}-2$:\ } Once again we take $X_\mu\rightarrow

1157: (X_\mu)_-$ as $x_6\rightarrow (m_i)_-$ but, from the other side, the

1158: matrices $X_\mu$ now have a simple pole at the boundary,

1159: %

1160: \be

1161: X_\mu\rightarrow \left(\begin{array}{cc} J_\mu/(x_6-m_i)+ Y_\mu &

1162: 0 \\ 0 & (X_\mu)_- \end{array}\right)\ \ \ \ \ {\rm as\ }

1163: x_6\rightarrow (m_i)_+

1164: \ee

1165: %

1166: where $J_\mu$ is the irreducible $(n_{i+1}-n_i)\times (n_{i+1}-n_i)$

1167: representation of $su(2)$, and $Y_\mu$ are now constant

1168: $(n_{i+1}-n_i)\times (n_{i+1}-n_i)$ matrices.

1169: 

1170: Case 2 above is usually described as a subset of Case 3 (with the

1171: one-dimensional irreducible $su(2)$ representation given by $J_\mu=0$).

1172: Here we have listed Case 2 separately since when we come to describe

1173: a similar construction for domain walls, only Case 1 and 2 above will

1174: appear. The conditions for $n_i<n_{i+1}$ were derived in \cite{chenwein}

1175: by starting with the impurity data \eqn{imp} and taking several monopoles

1176: to infinity. Obviously, for $n_i>n_{i+1}$, one imposes the same boundary

1177: conditions described above, only flipped in the $x_6$ direction.

1178: 

1179: The space of solutions to Nahm's equations, subject to the

1180: boundary conditions detailed above, is isomorphic to the monopole

1181: moduli space ${\cal M}_{\vg}$. Moreover, there exists a natural

1182: hyperK\"ahler metric on the solutions to Nahm's equations which

1183: can be shown to coincide with the Manton metric on the monopole

1184: moduli space. For the $\vg=\valpha_1+\valpha_2$ monopole, the

1185: metric on the associated Nahm data was computed in \cite{conell}

1186: and shown to give rise to the Euclidean Taub-NUT metric \eqn{tn}.

1187: For the $\vg=\sum_i\valpha_i$ monopoles, the corresponding

1188: computation was performed in \cite{murray}, resulting in the

1189: Lee-Weinberg-Yi metrics \cite{lwy2}.

1190: 

1191: 

1192: \section{D-Branes and Domain Walls}

1193: 

1194: In this section we would like to realize the domain walls that we

1195: described in Section 2 on the worldvolume of D-branes, mimicking

1196: Diaconescu's construction for monopoles.  From the resulting

1197: D-brane set-up we shall read off the world-volume dynamics of the

1198: domain walls to find that they are described by a truncated

1199: version of Nahm's equations \eqn{nahm}. Nahm's equations have also

1200: arisen as a description of domain walls in ${\cal N}=1^\star$

1201: theories \cite{boris}, although the relationship, if any, with the

1202: current work is unclear. Domain walls of the type described in

1203: Section 2 were previously embedded in D-branes in

1204: \cite{kinky,ohta} and several properties of the solitons were

1205: extracted (see in particular the latter reference). However, the

1206: worldvolume dynamics of the walls is difficult to determine in

1207: these set-ups and the relationship to magnetic monopoles obscured.

1208: 

1209: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1210: \begin{figure}[tbp]

1211: \begin{center}

1212: \epsfxsize=15cm\leavevmode\epsfbox{branes.eps}

1213: \end{center}

1214: \begin{center}

1215: \begin{minipage}{15cm}

1216:  \caption{\small

1217: The D-brane set-up for the $U(1)$ gauge theory with $N=3$ flavors. The

1218: vacuum is shown on the left; the elementary domain wall $\vg=\valpha_1$

1219: on the right.}

1220: \end{minipage}

1221: \end{center}

1222: \end{figure}

1223: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1224: 

1225: We start by constructing the theory with eight supercharges on the

1226: worldvolume of D-branes \cite{hw}. For definiteness we choose to

1227: build the ${\cal N}=2$, $d=3+1$ theory in IIA string theory

1228: although, by T-duality, we could equivalently work with any

1229: spacetime dimension\footnote{In fact, as explained in

1230: \cite{witten}, the overall $U(1)\subset U(k)$ is decoupled in

1231: the IIA brane set-up after lifting to M-theory. This effect will

1232: not concern us here.}. The construction is well known and is drawn

1233: in Figure 4. We suspend $k$ D4-branes between two NS5-branes,

1234: and insert a further $N$ D6-branes to play the role of the

1235: fundamental hypermultiplets. The worldvolume dimensions of the

1236: branes are

1237: %

1238: \be NS5: && 012345 \nn\\ D4: && 01236 \nn\\ D6: && 0123789 \nn\ee

1239: %

1240: The gauge coupling $e^2$ and the Higgs vev $v^2$ are encoded in

1241: the separation of the NS5-branes in the $x_6$ and $x_9$ directions

1242: respectively, while the masses $m_i$ are determined by the

1243: positions of the D6-branes in the $x_4$ direction (we choose the

1244: D6-branes to be coincident in the $x_5$ direction, corresponding

1245: to choosing all masses to be real).

1246: %

1247: \be \frac{1}{e^2} \sim \left.\frac{\Delta

1248: x_6}{l_sg_s}\right|_{NS5}\ \ \  \ ,\ \ \ \ v^2 \sim

1249: \left.\frac{\Delta x_9}{l_s^3g_s}\right|_{NS5}\ \ \ \ , \ \ \ \

1250: m_i\sim -\left.\frac{x_4}{l_s^2}\right|_{D6_i}\ee

1251: %

1252: After turning on the Higgs vev $v^2$, the D4-branes must split on

1253: the D6-branes in order to preserve supersymmetry. The S-rule

1254: \cite{hw} ensures that each D6-brane may play host to only a

1255: single D4-brane. In this manner a vacuum of the theory is chosen

1256: by picking $k$ out of the $N$ D6-branes on which the D4-branes

1257: end, in agreement with equation \eqn{set}.

1258: 

1259: The domain walls correspond to a configuration of D4-branes which

1260: start life at $x^3=-\infty$ in a vacuum configuration ${\Xi}_-$,

1261: and end up at $x^3=+\infty$ in a distinct vacuum $\Xi_+$. As is

1262: clear from Figure 4, as D4-branes walls interpolate in $x^1$, they

1263: must also move in both the $x^4$ direction and the $x^9$ direction

1264: \cite{hh}. The NS-branes and D6-branes are linked, meaning that a

1265: D4-brane is either created or destroyed as they pass the

1266: NS5-branes in the $x^6$ direction \cite{hw}. In the domain wall

1267: background, which of these possibilities occurs differs if we move

1268: the D6-branes to the left or right since the D4-brane charge

1269: varies from one end of the domain wall to the other.

1270: 

1271: As it stands, it is difficult to read off the dynamics of the

1272: D4-branes in Figure 4. However, we can make progress by taking the

1273: $e^2\rightarrow \infty$ limit, in which the two NS5-branes become

1274: coincident in the $x^6$ direction. After rotating our viewpoint,

1275: the system of branes now looks like the ladder configuration shown

1276: in Figure 5 (note that we have also rotated the branes relative to

1277: Figure 4, so the horizontal is the $x^4$ direction). We are left

1278: with a series of D4-branes, now with worldvolume $02349$,

1279: stretched between $N$ D6-branes, while simultaneously sandwiched

1280: between two NS5-branes. Following these manoeuvres, one finds that

1281: the domain wall $\vg=\sum_in_i\valpha_i$ results in $n_i$

1282: D4-branes stretched between the $i^{\rm th}$ and $(i+1)^{\rm th}$

1283: D6-branes (counting from the top, since we

1284: %The figure indicates the opposite: m_i < m_{i+1}

1285: have chosen the ordering $m_i > m_{i+1}$.

1286: 

1287: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1288: \begin{figure}[tbp]

1289: \begin{center}

1290: \epsfxsize=15cm\leavevmode\epsfbox{dwbranes.eps}

1291: \end{center}

1292: \begin{center}

1293: \begin{minipage}{15cm}

1294:  \caption{\small

1295: The D-brane set-up for the $U(2)$ gauge theory with $N=6$ flavors. The

1296: maximal

1297: $\vg=\valpha_1+\valpha_5+2(\valpha_2+\valpha_3+\valpha_4)$ domain wall

1298: is shown.}

1299: \end{minipage}

1300: \end{center}

1301: \end{figure}

1302: %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%

1303: 

1304: It may be worth describing how the domain wall charges arise directly

1305: in the set-up of Figure

1306: % changed label

1307: 5. We start in a chosen vacuum $\Xi_-$, denoted

1308: by placing $k$ pairs of white dots on $N$ distinct D6-branes, as

1309: shown in the figure. A domain wall arises every time a pair of dots

1310: is proceeds to another D6-brane, dragging a D4-brane behind it like

1311: clingwrap. The S-rule translates to the fact that two pairs of dots may not

1312: simultaneously lie on the same D6-brane. The final vacuum $\Xi_+$ is

1313: denoted by the black dots in the figure and the domain wall charges

1314: $n_i$ are given by the number of times a D4-brane has been pulled between

1315: the $i^{\rm th}$ and $(i+1)^{\rm th}$ D6-branes.

1316: 

1317: \subsection{Domain Wall Dynamics}

1318: 

1319: We are now in a position to read off the dynamics of the domain

1320: walls. In the absence of the NS5-branes, the D4-branes would

1321: stretch to infinity in the $x_9$ direction, and the resulting

1322: D-brane set-up in Figure 5 is T-dual to the monopoles in Figure 3.

1323: The presence of the NS5-branes projects out half the degrees of

1324: freedom of the monopoles, leaving a simple linear set of

1325: equations. In each segment $m_i\leq x_4\leq m_{i+1}$ the domain

1326: walls are described by two $n_i\times n_i$ matrices $X_3$ and

1327: $A_4$ satisfying

1328: %

1329: \be \frac{dX_3}{dx_4}-i[A_4,X_3]=0 \label{dwnahm}\ee

1330: %

1331: modulo $U(n_i)$ gauge transformations acting on the interval $m_i\leq

1332: x_4\leq m_{i+1}$, and vanishing  at the boundaries. As in the case of

1333: monopoles, the interactions between neighbouring segments depends on the

1334: relative size of the matrices:

1335: 

1336: \underline{\ $n_i=n_{i+1}$:\ } Again, the $U(n_i)$ gauge symmetry is

1337: extended to the interval $m_i\leq x_4\leq m_{i+2}$ and an impurity is

1338: added to the right-hand-side of Nahm's equations, which now read

1339: %

1340: \be \frac{dX_3}{dx_4}-i[A_4,X_3]=

1341: \pm\psi\psi^\dagger\delta(x_4-m_{i+1}) \label{dwimp}\ee

1342: %

1343: where the impurity degree of freedom $\psi$ transforms in the

1344: fundamental representation of the $U(n_i)$ gauge group, ensuring

1345: the combination $\psi\psi^\dagger$ is a $n_i\times n_i$ matrix

1346: transforming, like $X_3$, in the adjoint representation. These

1347: $\psi$ degrees of freedom are chiral multiplets which survive the

1348: NS5-brane projection. We shall see shortly that the choice of

1349: $\pm$ sign will dictate the relative ordering of the domain walls

1350: along the $x_3$ direction.

1351: 

1352: \underline{\ $n_i=n_{i+1}-1$:\ } In this case $X_3\rightarrow

1353: (X_3)_-$, an $n_i\times n_i$ matrix, as $x_4\rightarrow (m_i)_-$

1354: from the left. To the right of $m_i$, $X_3$ is a $(n_i+1)\times

1355: (n_i+1)$ matrix obeying

1356: %

1357: \be X_3\rightarrow \left(\begin{array}{cc} y & a^\dagger \\ a &

1358: (X)_-

1359: \end{array}\right)\ \ \ \ \ {\rm as\ } x_4\rightarrow (m_i)_+

1360: \label{bc}\ee

1361: %

1362: where $y_\mu\in\R$ and each $a_\mu$ is a complex $n_i$-vector. The obvious analog

1363: of this boundary condition holds when $n_i=n_{i+1}+1$.

1364: 

1365: These boundary conditions obviously descend from the original Nahm

1366: boundary conditions for monopoles. Just as the space of Nahm data

1367: is isomorphic to the moduli space of magnetic monopoles, we

1368: conjecture that the moduli space of linearized Nahm data described

1369: above is isomorphic to the moduli space of domain walls. We shall

1370: shortly show that it indeed captures the most relevant aspect of

1371: domain walls: their ordering. In fact, the linearized Nahm

1372: equations \eqn{dwnahm} are rather trivial to solve. We first

1373: employ the $\prod_iU(n_i)$ gauge transformations to make

1374: $A_4(x_4)$ a constant in each interval $m_i\leq x_4\leq m_{i+1}$.

1375: This can be achieved by first diagonalizing $A_4$, and

1376: subsequently acting with the $U(1)^{n_i}$ transformation

1377: $A_4\rightarrow A_4-\partial_4\alpha$ where, in each segment

1378: $m_i\leq x_4\leq m_{i+1}$, $\alpha$ is given by

1379: %

1380: \be \alpha(x_4)=\int_{m_i}^{x_4}\ A_4(x^\prime_4)\,dx_4^\prime -\left[\int_{m_i}^{m_{i+1}}

1381: A_4(x^\prime_4)\,dx^\prime_4\right]\frac{m_i-x_4}{m_i-m_{i+1}}

1382: \label{gt}\ee

1383: %

1384: which has the property that $\alpha(m_i)=\alpha(m_{i+1})=0$.

1385: Further gauge transformations with non-zero winding on the

1386: interval ensure that $A_4$ is periodic, with each eigenvalue lying

1387: in $A_4\in [0,2\pi /(m_i-m_{i+1}))$. These $N-1$ ``Wilson lines''

1388: will play the role of the phases associated to domain wall system.

1389: Note that when $n_i=n_{i+1}$, the above choice of gauge leaves a

1390: residual $U(n_i)$ gauge symmetry acting only on the chiral

1391: impurity $\psi$. In this gauge we can now easily integrate

1392: \eqn{dwnahm} in each interval,

1393: %

1394: \be X_3(x_4)=e^{iA_4x_4}\,\hat{X}_3\,e^{-iA_4x_4} \ee

1395: %

1396: where the eigenvalues of $X_3$ are independent of $x_4$ in each

1397: interval. We identify these $n_i$ eigenvalues with the positions

1398: of the $n_i$ $\valpha_i$ elementary domain walls.

1399: 

1400: We are now in a position to derive the linearized Nahm equations

1401: \eqn{dwnahm} from the original Nahm equations \eqn{nahm} in terms

1402: of a fixed point set of a $\hat{U}(1)$ action. Consider first the

1403: action of the $U(1)_R\subset SU(2)_R$ isometry on the Nahm data,

1404: which rotates $X_1$ and $X_2$ while leaving $X_3$ fixed. This

1405: rotation also acts on the impurity

1406: $\omega=(\psi,\tilde{\psi}^\dagger)$ by $(\psi,

1407: \tilde{\psi})\rightarrow e^{i\alpha}(\psi,\tilde{\psi})$. To

1408: retain half of the impurities for the domain wall equations

1409: \eqn{dwimp}, we need to compensate for this transformation with

1410: the residual $U(1)\subset U(n_i)$ transformation acting on the

1411: appropriate impurity $\omega$ by $\omega\rightarrow

1412: e^{i\beta}\omega$. By choosing $\beta=\pm\alpha$ we can pick a

1413: $\hat{U}(1)$ action which leaves either the $\psi$ or the

1414: $\tilde{\psi}$ impurity invariant. Which we choose to save is

1415: correlated with the choice of minus sign in \eqn{dwimp} which, in

1416: turn, dictates the ordering of neighbouring domain walls as we

1417: shall now demonstrate.

1418: 

1419: To summarize, we have shown the that description of domain wall

1420: dynamics \eqn{dwnahm} arises from the fixed point of a

1421: $\hat{U}(1)$ on the original Nahm equations \eqn{nahm}. This

1422: action descends to a $\hat{U}(1)$ isometry on the monopole moduli

1423: space ${\cal M}_{\vg}$, the fixed points of which coincide with

1424: the domain wall moduli space ${\cal W}_{\vg}$. A physical explanation

1425: for this correspondence follows along the lines of \cite{stillme}:

1426: in theories in the Higgs phase, confined magnetic monopoles with charge

1427: $\vg$ exist, emitting $k$ multiple vortex strings. When

1428: these vortex strings coincide, the worldvolume theory is of the form

1429: described in Section 2 \cite{vib} and the monopoles appear as charge

1430: $\vg$ kinks.

1431: 

1432: 

1433: \subsection{The Ordering of Domain Walls Revisited}

1434: 

1435: As explained in Section 2, in contrast to monopoles, domain walls

1436: must satisfy a specific ordering on the $x^3$ line. We will now

1437: show that this ordering is encoded in the boundary conditions

1438: described above. Suppose first that $n_i=n_{i+1}$. The positions

1439: of the $\valpha_i$ domain walls are given by the eigenvalues of

1440: $X_3$ restricted to the interval $m_i\leq x_4\leq m_{i+1}$. Let us

1441: denote this matrix as $X_3^{(i)}$ and the eigenvalues as

1442: $\lambda^{(i)}_m$, where $m=1,\ldots n_i$.  The impurity

1443: \eqn{dwimp} relates the two sets of eigenvalues by the jumping

1444: condition

1445: %

1446: \be X_3^{(i+1)}=X_3^{(i)}+\psi\psi^\dagger \label{jump}\ee

1447: %

1448: where we have chosen the positive sign for definiteness.  However,

1449: from the discussion in Section 2 (see, in particular, figure 3) we

1450: know that the domain walls cannot have arbitrary position but must

1451: be interlaced,

1452: %

1453: \be

1454: \lambda^{(i)}_1\leq \lambda^{(i+1)}_1\leq\lambda^{(i)}_2\leq\ldots\leq\lambda^{(i+1)}_{n_i-1}

1455: \leq\lambda^{(i)}_{n_i}\leq\lambda^{(i+1)}_{n_i}

1456: \label{order}\ee

1457: %

1458: We will now show that the ordering of domain walls \eqn{order} follows from

1459: the impurity jumping condition \eqn{jump}.

1460: 

1461: To see this, consider firstly the situation in which

1462: $\psi^\dagger\psi\ll \Delta\lambda^{(i)}_{m}$ so that the matrix

1463: $\psi\psi^\dagger$ may be treated as a small perturbation of

1464: $X_3^{(i)}$. The positivity of $\psi\psi^\dagger$ ensures that

1465: each $\lambda_m^{(i+1)}\geq \lambda_m^{(i)}$. Moreover, it is

1466: simple to show that the $\lambda_m^{(i+1)}$ increase monotonically

1467: with $\psi^\dagger\psi$. This leaves us to consider the other

1468: extreme, in which $\psi^\dagger\psi\rightarrow \infty$. It this

1469: limit $\psi$ becomes one of the eigenvectors of $X_3^{(i+1)}$ with

1470: eigenvalue $\lambda^{(i+1)}_{n_i}=\psi^\dagger\psi\rightarrow

1471: \infty$ which reflects the fact that this limit corresponds to the

1472: situation in which the last domain wall is taken to infinity. What

1473: we want to show is that the remaining $(n_i-1)$ $\valpha_{i+1}$

1474: domain walls are trapped between the $n_i$ $\valpha_i$ domain

1475: walls as depicted in Figure 3. Define the $n_i\times n_i$

1476: projection operator

1477: %

1478: \be

1479: P=1-\hat{\psi}\hat{\psi}^\dagger

1480: \ee

1481: %

1482: where $\hat{\psi}=\psi/\sqrt{\psi^\dagger\psi}$. The positions of

1483: the remaining $(n_i-1)$ $\valpha_{i+1}$ domain walls are given by

1484: the (non-zero) eigenvalues of $PX_3^{(i)}P$. We must show that,

1485: given a rank $n$ hermitian matrix $X$, the eigenvalues of $PXP$

1486: are trapped between the eigenvalues of $X$. This elementary

1487: property of hermitian matrices can be seen as follows:

1488: %

1489: \be

1490: \det(PXP-\mu)&=&\det(XP-\mu) \nn\\

1491: &=&\det(X-\mu-X\hat{\psi}\hat{\psi}^\dagger)\nn\\ &=&\det(X-\mu)

1492: \det(1-(X-\mu)^{-1}X\hat{\psi}\hat{\psi}^\dagger)\nn\ee

1493: %

1494: Since $\hat{\psi}\hat{\psi}^\dagger$ is rank one, we can write this as

1495: %

1496: \be

1497: \det(PXP-\mu)&=&

1498: \det(X-\mu)\,[1-\Tr((X-\mu)^{-1}X\hat{\psi}\hat{\psi}^\dagger)]\nn\\

1499: &=& -\mu\,\det(X-\mu)\,\Tr((X-\mu)^{-1}\hat{\psi}\hat{\psi}^\dagger)

1500: \nn\\ &=& -\mu\left[\prod_{m=1}^n(\lambda_m-\mu)\right]\,\left[\sum_{m=1}^n

1501: \frac{|\hat{\psi}_m|^2}{\lambda_m-\mu}\right]

1502: \ee

1503: where $\hat{\psi}_m$ is the $m^{\rm th}$ component of the vector $\psi$.

1504: We learn that $PXP$ has one zero eigenvalue while, if the eigenvalues

1505: $\lambda_m$ of $X$ are distinct, then the eigenvalues of $PXP$ lie at the

1506: roots the function

1507: %

1508: \be R(\mu)=\sum_{m=1}^n\frac{|\hat{\psi}_m|^2}{\lambda_m-\mu}\ee

1509: %

1510: The roots of $R(\mu)$ indeed lie between the eigenvalues $\lambda_m$. This

1511: completes the proof that the impurities \eqn{dwimp} capture the correct ordering

1512: of the domain walls.

1513: 

1514: The same argument shows that the boundary condition \eqn{bc} gives rise to the

1515: correct ordering of domain walls when $n_{i+1}=n_i+1$, with the $\valpha_{i}$

1516: domain walls interlaced between the $\valpha_{i+1}$ domains walls. Indeed, it

1517: is not hard to show that \eqn{bc} arises from \eqn{dwimp} in the limit that

1518: one of the domain walls is taken to infinity.

1519: 

1520: 

1521: %\section{Confined Monopoles}

1522: %

1523: %In this final section we would like to describe the relationship between

1524: %confined magnetic monopoles and kinks (or domain walls) on the worldvolume

1525: %of a vortex flux tubes. This relationship was described in detail in

1526: %\cite{stillme,shif,meami} for confined monopoles on a single ($k=1$)

1527: %flux tube. The extension to monopoles threaded on multiple flux tubes is

1528: %somewhat trivial and here we simply sketch the ideas with a broad brush.

1529: %

1530: %The monopoles  described in Section 3 exist in theories which lie

1531: %in the Coulomb phase, where the $SU(N)$ gauge symmetry is broken

1532: %to the maximal torus $U(1)^{N-1}$. Here we consider the same

1533: %monopoles when the theory lies in a phase with fully broken gauge

1534: %symmetry. To achieve this, we first extend the gauge group to

1535: %$U(N)$ and, then add $N$ scalars, each transforming in the

1536: %fundamental representation. Let's call them $u^a_{\ i}$, where

1537: %$a=1,\ldots, N$ is the color index and $i=1,\ldots N$ is the

1538: %flavor index. We consider the Lagrangian

1539: %%

1540: %\be {\cal L} ={\Tr} \left[\frac{1}{4e^2} F_{\mu\nu}F^{\mu\nu}

1541: %+\frac{1}{2e^2}|{\cal D}_\mu\phi|^2 +\frac{e^2}{2}(u_i\!\otimes

1542: %\!u_i^\dagger-\zeta^2)^2\right] +\sum_{i=1}^{N}\left[|{\cal D}_\mu

1543: %u_i|^2 +u_i^\dagger(\phi-m_i)^2u_i \right]\nn\ee

1544: %%

1545: %The astute reader may notice that this is rather similar to the

1546: %Lagrangian of Section 2. However, since we have chosen the number

1547: %of flavors to be equal to the number of colors above, the theory

1548: %has a unique ground state $\phi={\rm diag}(m_1,\ldots, m_{N})$ and

1549: %$u^a_{\ i}=\zeta\delta^a_{\ i}$ and does not admit domain wall

1550: %solutions. Nor does it admit monopole solutions of the form

1551: %\eqn{monbog}. Physically the reason is clear: the theory does not

1552: %have any massless photons. However, as shown in \cite{stillme},

1553: %the monopole does survive as a BPS solution of the theory, with

1554: %its flux now emitted in two, collimated tubes. These tubes are

1555: %vortex strings. The Bogomoln'yi equations describing the

1556: %configuration in which the flux tubes are emitted in the $x^3$

1557: %direction are,

1558: %

1559: %\be &B_3={\cal D}_3\phi+e^2(\sum_{i=1}^Nu_iu_i^\dagger-\zeta^2)\ \

1560: %\ ,\ \ \

1561: %B_1={\cal D}_1\phi\ \ \ ,\ \ \ B_2={\cal D}_2\phi&\nn\\

1562: %&{\cal D}_3u_i=(\phi-m_i)u_i\ \ \ , \ \ \ {\cal D}_1u_i=i{\cal

1563: %D}_2u_i& \label{mybog}\ee

1564: %%

1565: %At this point, we change perspective. We now describe this

1566: %confined magnetic monopole from the perspective of the flux tube.

1567: %This flux tube is simply the vortex string, satisfying a subset of

1568: %\eqn{mybog}: $B_3=e^2(\sum_{i=1}^Nu_iu_i^\dagger-\zeta^2)$ and

1569: %${\cal D}_1u_i=i{\cal D}_2u_i$. In fact, there are $N$ different

1570: %types of vortex strings, each with magnetic field lying in a

1571: %different diagonal component of the $U(N)$ gauge group. Consider

1572: %$k$ parallel vortex strings, each of arbitrary type. The $d=1+1$

1573: %dynamics of these vortex strings was described in \cite{vib} and

1574: %is given by an auxiliary $U(k)$ gauge theory with field strength

1575: %$G_{03}$, a real adjoint scalar $\sigma$, a complex adjoint scalar

1576: %$Z=X_2+iX_3$ and $N$ fundamental scalars $q_i$, interacting via

1577: %the worldsheet Lagrangian

1578: %%

1579: %\be {\cal L}_{\rm vortex} &=&{\Tr} \left[\frac{1}{2g^2} G_{01}^2+\frac{1}{2g^2}|{\cal D}_\alpha\sigma|^2

1580: %+|{\cal D}_\alpha Z|^2\right]+\sum_{i=1}^{N}|{\cal D}_\alpha q_i|^2\label{vortex}\\&&

1581: %+\ \frac{g^2}{2}\Tr(\sum_{i=1}^Nq_i\!\otimes\!q_i^\dagger+[Z,Z^\dagger]-r)^2

1582: %+\Tr|[\sigma,Z]|^2+\sum_{i=1}^{N}q_i^\dagger(\sigma-m_i)^2q_i

1583: %\nn\ee

1584: %

1585: %where $r=2\pi/e^2$, and the auxiliary gauge coupling is to be

1586: %taken to infinity: $g^2\rightarrow\infty$. This Lagrangian has

1587: %several disconnected branches of vacua for all $k$ and $N$. For

1588: %example, in the case of $k=1$, there are $N$ vacua given by

1589: %$\sigma=m_i$ and $q_j=\sqrt{r}\delta_{ij}$. These correspond to

1590: %the $N$ different vortex strings described above. The eigenvalues

1591: %of $Z$ are identified with the positions of the vortex strings. To

1592: %make contact with the domain walls described in Section 2, we must

1593: %first place all $k$ strings on top of each other by setting $Z=0$.

1594: %This is only possible while preserving supersymmetry for $k\leq N$

1595: %strings so we restrict to this case. The Lagrangian \eqn{vortex}

1596: %now reduces to that discussed in Section 2.

1597: %

1598: %The identification of the domain walls in the vortex theory

1599: %\eqn{vortex} with the confined magnetic monopoles \eqn{mybog}

1600: %follows from studying the quantum numbers of each object

1601: %\cite{stillme}. Since the different vacua correspond to vortex

1602: %strings carrying magnetic flux in different $U(1)$ subgroups of

1603: %$U(N)$, the kink which takes us from one vacua to the other must

1604: %be a source for magnetic flux. It is not hard to see that we the

1605: %confined magnetic monopole of charge $\vec{g}$ becomes a kink on

1606: %the vortex worldsheet in the corresponding topologial sector

1607: %$\vg$. Moreover, the mass of the kink and the monopole coincide;

1608: %from the formula \eqn{tvg} we have

1609: %%

1610: %\be

1611: %M_{\rm kink}=r\vg\cdot\vec{m}=\frac{2\pi}{e^2}\vec{m}\cdot\vg = M_{\rm mono}

1612: %\ee

1613: %%

1614: %In summary, the magnetic monopoles of Section 3 survive in

1615: %the Higgs phase of the theory, confined to move on a line as beads

1616: %threaded on vortex flux tubes. From the perspective of the coincident

1617: %flux tubes, the monopoles are seen as domain walls of the type described in

1618: %Section 2. This provides a physical explanation for the relationship between the

1619: %moduli spaces of monopoles and domain walls that we have described in this

1620: %paper.

1621: 

1622: \section*{Acknowledgement}

1623: AH is supported in part by the CTP and LNS of MIT, DOE contract

1624: $\#$DE-FC02-94ER40818, NSF grant PHY-00-96515, the BSF

1625: American-Israeli Bi-national Science Foundation and a DOE OJI

1626: Award.  DT is supported by the Royal Society.

1627: 

1628: \begin{thebibliography}{99}

1629: 

1630: \small

1631: \parskip=0pt plus 2pt

1632: 

1633: 

1634: \bibitem{edpaul} E.~R.~C.~Abraham and P.~K.~Townsend,

1635:   ``{\it Q kinks},''

1636:   Phys.\ Lett.\ B {\bf 291}, 85 (1992).;

1637:   %%CITATION = PHLTA,B291,85;%%

1638: ``{\it More on Q kinks: A (1+1)-dimensional analog of dyons},''

1639:   Phys.\ Lett.\ B {\bf 295}, 225 (1992).

1640:   %%CITATION = PHLTA,B295,225;%%

1641: 

1642: \bibitem{nick} N. Dorey, ``{\it The BPS spectra of two-dimensional supersymmetric gauge theories with

1643:   twisted mass terms},''

1644:   JHEP {\bf 9811}, 005 (1998)

1645:   [arXiv:hep-th/9806056]. \\

1646:  N.~Dorey, T.~J.~Hollowood and D.~Tong,

1647:   ``{\it The BPS spectra of gauge theories in two and four dimensions},''

1648:   JHEP {\bf 9905}, 006 (1999)

1649:   [arXiv:hep-th/9902134].

1650:   %%CITATION = HEP-TH 9902134;%%

1651:   %%CITATION = HEP-TH 9806056;%%

1652: 

1653: \bibitem{stillme} D.~Tong,

1654:   ``{\it Monopoles in the Higgs phase},''

1655:   Phys.\ Rev.\ D {\bf 69}, 065003 (2004)

1656:   [arXiv:hep-th/0307302].

1657:   %%CITATION = HEP-TH 0307302;%%

1658: 

1659: \bibitem{bead}

1660:   M.~Hindmarsh and T.~W.~B.~Kibble,

1661:   ``{\it Beads On Strings},''

1662:   Phys.\ Rev.\ Lett.\  {\bf 55}, 2398 (1985).

1663:   %%CITATION = PRLTA,55,2398;%%

1664: 

1665: 

1666: \bibitem{shif}   M.~Shifman and A.~Yung,

1667:   ``{\it Non-Abelian string junctions as confined monopoles},''

1668:   Phys.\ Rev.\ D {\bf 70}, 045004 (2004)

1669:   [arXiv:hep-th/0403149].

1670:   %%CITATION = HEP-TH 0403149;%%

1671: 

1672: \bibitem{meami}   A.~Hanany and D.~Tong,

1673:   ``{\it Vortex strings and four-dimensional gauge dynamics},''

1674:   JHEP {\bf 0404}, 066 (2004)

1675:   [arXiv:hep-th/0403158].

1676:   %%CITATION = HEP-TH 0403158;%%

1677: 

1678: \bibitem{auzzi}  R.~Auzzi, S.~Bolognesi and J.~Evslin,

1679:   ``{\it Monopoles can be confined by 0, 1 or 2 vortices}''

1680:   JHEP {\bf 0502}, 046 (2005)

1681:   [arXiv:hep-th/0411074].

1682:   %%CITATION = HEP-TH 0411074;%%

1683: 

1684: 

1685: \bibitem{o1}

1686:  M.~A.~C.~Kneipp,

1687:   ``{\it Color superconductivity, Z(N) flux tubes and monopole confinement in

1688:   deformed N = 2* super Yang-Mills theories},''

1689:   Phys.\ Rev.\ D {\bf 69}, 045007 (2004)

1690:   [arXiv:hep-th/0308086].

1691:   %%CITATION = HEP-TH 0308086;%%

1692: \bibitem{o2}  R.~Auzzi, S.~Bolognesi, J.~Evslin and K.~Konishi,

1693:   ``{\it Nonabelian monopoles and the vortices that confine them},''

1694:   Nucl.\ Phys.\ B {\bf 686}, 119 (2004)

1695:   [arXiv:hep-th/0312233].

1696:   %%CITATION = HEP-TH 0312233;%%

1697: \bibitem{o3}  V.~Markov, A.~Marshakov and A.~Yung,

1698:   ``{\it Non-Abelian vortices in N = 1* gauge theory},''

1699:   Nucl.\ Phys.\ B {\bf 709}, 267 (2005)

1700:   [arXiv:hep-th/0408235].

1701:   %%CITATION = HEP-TH 0408235;%%}

1702: \bibitem{o4} A.~Gorsky, M.~Shifman and A.~Yung,

1703:  ``{\it Non-Abelian Meissner effect in Yang-Mills theories at weak coupling},''

1704:   Phys.\ Rev.\ D {\bf 71}, 045010 (2005)

1705:   [arXiv:hep-th/0412082].

1706:   %%CITATION = HEP-TH 0412082;%%

1707: \bibitem{o5}  A.~Mironov, A.~Morozov and T.~N.~Tomaras,

1708:   ``{\it On the need for phenomenological theory of P-vortices or does spaghetti

1709:   confinement pattern admit condensed-matter analogies?},''

1710:   arXiv:hep-th/0503212.

1711:   %%CITATION = HEP-TH 0503212;%%

1712: \bibitem{o6}  S.~Bolognesi and J.~Evslin,

1713:   ``{\it Stable vs unstable vortices in SQCD},''

1714:   arXiv:hep-th/0506174.

1715:   %%CITATION = HEP-TH 0506174;%%

1716: 

1717: 

1718: \bibitem{vib} A. Hanany and D. Tong, "{\it Vortices, Instantons

1719: and Branes}'', JHEP {\bf 0307}, 037 (2003)

1720:   [arXiv:hep-th/0306150].

1721:   %%CITATION = HEP-TH 0306150;%%

1722: 

1723: \bibitem{web} M.~Eto, Y.~Isozumi, M.~Nitta, K.~Ohashi and N.~Sakai,

1724:   ``{\it Webs of walls},''

1725:   arXiv:hep-th/0506135.

1726:   %%CITATION = HEP-TH 0506135;%%

1727: 

1728: \bibitem{14}    K.~Lee and H.~U.~Yee,

1729:   ``{\it New BPS Objects in N=2 Supersymmetric Gauge Theories},''

1730:   arXiv:hep-th/0506256. \\

1731:   %%CITATION = HEP-TH 0506256;%%

1732:  M.~Eto, Y.~Isozumi, M.~Nitta and K.~Ohashi,

1733:   ``{\it $\ft12$, $\ft14$ and $\ft18$ BPS Equations in SUSY Yang-Mills-Higgs Systems: Field

1734:   Theoretical Brane Configurations},''

1735:   arXiv:hep-th/0506257.

1736:   %%CITATION = HEP-TH 0506257;%%

1737: 

1738: 

1739: \bibitem{me}   D.~Tong,

1740:   ``{\it The moduli space of BPS domain walls},''

1741:   Phys.\ Rev.\ D {\bf 66}, 025013 (2002)

1742:   [arXiv:hep-th/0202012].

1743:   %%CITATION = HEP-TH 0202012;%%

1744: 

1745: \bibitem{isozumi}  Y.~Isozumi, M.~Nitta, K.~Ohashi and N.~Sakai,

1746:   ``{\it Construction of non-Abelian walls and their complete moduli space},''

1747:   Phys.\ Rev.\ Lett.\  {\bf 93}, 161601 (2004)

1748:   [arXiv:hep-th/0404198]. \\  ``{\it All exact solutions of a 1/4 BPS equation},''

1749:   Phys.\ Rev.\ D {\bf 71}, 065018 (2005)

1750:   [arXiv:hep-th/0405129].

1751:   %%CITATION = HEP-TH 0405129;%%

1752:   %%CITATION = HEP-TH 0404198;%%

1753: 

1754: \bibitem{isonon}   Y.~Isozumi, M.~Nitta, K.~Ohashi and N.~Sakai,

1755:   ``{\it Non-Abelian walls in supersymmetric gauge theories},''

1756:   Phys.\ Rev.\ D {\bf 70}, 125014 (2004)

1757:   [arXiv:hep-th/0405194].

1758:   %%CITATION = HEP-TH 0405194;%%

1759: 

1760: 

1761: \bibitem{finite}  Y.~Isozumi, K.~Ohashi and N.~Sakai,

1762:   ``{\it Exact wall solutions in 5-dimensional SUSY QED at finite coupling}''

1763:   JHEP {\bf 0311}, 060 (2003)

1764:   [arXiv:hep-th/0310189]. \\

1765:   %%CITATION = HEP-TH 0310189;%%

1766:   N.~Sakai and Y.~Yang,

1767:   ``{\it Moduli sapce of BPS walls in supersymmetric gauge theories},''

1768:   arXiv:hep-th/0505136.

1769:   %%CITATION = HEP-TH 0505136;%%

1770: 

1771: 

1772: 

1773: \bibitem{boojum} N. Sakai and D. Tong, ``{\it Monopoles, Vortices, Domain Walls

1774: and D-Branes: The Rules of Interaction}''  JHEP {\bf 0503}, 019 (2005)

1775:   [arXiv:hep-th/0501207].

1776:   %%CITATION = HEP-TH 0501207;%%

1777: 

1778: \bibitem{lee}

1779:   K.~S.~M.~Lee,

1780:   ``{\it An index theorem for domain walls in supersymmetric gauge theories},''

1781:   Phys.\ Rev.\ D {\bf 67}, 045009 (2003)

1782:   [arXiv:hep-th/0211058].

1783:   %%CITATION = HEP-TH 0211058;%%

1784: 

1785: 

1786: \bibitem{mmotw}   D.~Tong,

1787:   ``{\it Mirror mirror on the wall: On two-dimensional black holes and Liouville

1788:   theory},''

1789:   JHEP {\bf 0304}, 031 (2003)

1790:   [arXiv:hep-th/0303151].

1791:   %%CITATION = HEP-TH 0303151;%%

1792: 

1793: 

1794: \bibitem{gno}   P.~Goddard, J.~Nuyts and D.~I.~Olive,

1795:   ``{\it Gauge Theories And Magnetic Charge},''

1796:   Nucl.\ Phys.\ B {\bf 125}, 1 (1977).

1797:   %%CITATION = NUPHA,B125,1;%%

1798: 

1799: 

1800: \bibitem{erick} E.~J.~Weinberg,

1801:   ``{\it Fundamental Monopoles And Multi - Monopole Solutions For Arbitrary Simple

1802:   Gauge Groups},''

1803:   Nucl.\ Phys.\ B {\bf 167}, 500 (1980).

1804:   %%CITATION = NUPHA,B167,500;%%

1805: 

1806: \bibitem{manton} N.~S.~Manton,

1807:   ``{\it A Remark On The Scattering Of BPS Monopoles},''

1808:   Phys.\ Lett.\ B {\bf 110}, 54 (1982).

1809:   %%CITATION = PHLTA,B110,54;%%

1810: 

1811: 

1812: \bibitem{ah} M. Atiyah and N. Hitchin, ``{\it The Geometry and Dynamics

1813: of Magnetic Monopoles}'', Princteon University Press (1988).

1814: 

1815: \bibitem{conell} S. A. Connell, ``{\it The Dynamics of the $SU(3)$ Charge $(1,1)$ Magnetic

1816: Monopole},'', University of South Australia Preprint.

1817: 

1818: \bibitem{gl} J.~P.~Gauntlett and D.~A.~Lowe,

1819:   ``{\it Dyons and S-Duality in N=4 Supersymmetric Gauge Theory},''

1820:   Nucl.\ Phys.\ B {\bf 472}, 194 (1996)

1821:   [arXiv:hep-th/9601085].

1822:   %%CITATION = HEP-TH 9601085;%%

1823: 

1824: 

1825: 

1826: \bibitem{lwy1} K. Lee, E. Weinberg and P. Yi,   ``{\it Electromagnetic Duality and $SU(3)$

1827: Monopoles},''

1828:   Phys.\ Lett.\ B {\bf 376}, 97 (1996)

1829:   [arXiv:hep-th/9601097].

1830:   %%CITATION = HEP-TH 9601097;%%

1831: 

1832: 

1833: \bibitem{lwy2} K. Lee, E. Weinberg and P. Yi,

1834: ``{\it The Moduli Space of Many BPS Monopoles for Arbitrary Gauge Groups},''

1835:   Phys.\ Rev.\ D {\bf 54}, 1633 (1996)

1836:   [arXiv:hep-th/9602167].

1837:   %%CITATION = HEP-TH 9602167;%%

1838: 

1839: \bibitem{21} C.~Houghton, P.~W.~Irwin and A.~J.~Mountain,

1840:   ``{\it Two monopoles of one type and one of another},''

1841:   JHEP {\bf 9904}, 029 (1999)

1842:   [arXiv:hep-th/9902111].

1843:   %%CITATION = HEP-TH 9902111;%%

1844: 

1845: \bibitem{nahm}

1846:   W.~Nahm,

1847:   ``{\it A Simple Formalism For The BPS Monopole},''

1848:   Phys.\ Lett.\ B {\bf 90}, 413 (1980).

1849:   %%CITATION = PHLTA,B90,413;%%

1850: 

1851: 

1852: \bibitem{diac} D.~E.~Diaconescu,

1853:   ``{\it D-branes, monopoles and Nahm equations},''

1854:   Nucl.\ Phys.\ B {\bf 503}, 220 (1997)

1855:   [arXiv:hep-th/9608163].

1856:   %%CITATION = HEP-TH 9608163;%%

1857: 

1858: 

1859: \bibitem{hub}

1860:   J.~Hurtubise and M.~K.~Murray,

1861:   ``{\it On The Construction Of Monopoles For The Classical Groups},''

1862:   Commun.\ Math.\ Phys.\  {\bf 122}, 35 (1989).

1863:   %%CITATION = CMPHA,122,35;%%

1864: 

1865: 

1866: \bibitem{hw}  A.~Hanany and E.~Witten,

1867:   ``{\it Type IIB superstrings, BPS monopoles, and three-dimensional gauge

1868:   dynamics},''

1869:   Nucl.\ Phys.\ B {\bf 492}, 152 (1997)

1870:   [arXiv:hep-th/9611230].

1871:   %%CITATION = HEP-TH 9611230;%%

1872: 

1873: \bibitem{kapset}  A.~Kapustin and S.~Sethi,

1874:  ``{\it The Higgs branch of impurity theories},''

1875:   Adv.\ Theor.\ Math.\ Phys.\  {\bf 2}, 571 (1998)

1876:   [arXiv:hep-th/9804027].

1877:   %%CITATION = HEP-TH 9804027;%%

1878: 

1879: 

1880: \bibitem{tsimpis}  D.~Tsimpis,

1881:   ``{\it Nahm equations and boundary conditions},''

1882:   Phys.\ Lett.\ B {\bf 433}, 287 (1998)

1883:   [arXiv:hep-th/9804081].

1884:   %%CITATION = HEP-TH 9804081;%%

1885: 

1886: 

1887: \bibitem{chenwein} X.~G.~Chen and E.~J.~Weinberg,

1888:   ``{\it ADHMN boundary conditions from removing monopoles},''

1889:   Phys.\ Rev.\ D {\bf 67}, 065020 (2003)

1890:   [arXiv:hep-th/0212328].

1891:   %%CITATION = HEP-TH 0212328;%%

1892: 

1893: 

1894: \bibitem{murray} M.~K.~Murray,

1895:   ``{\it A note on the (1, 1,..., 1) monopole metric},''

1896:   J.\ Geom.\ Phys.\  {\bf 23}, 31 (1997)

1897:   [arXiv:hep-th/9605054].

1898:   %%CITATION = HEP-TH 9605054;%%

1899: 

1900: \bibitem{boris}   C.~Bachas, J.~Hoppe and B.~Pioline,

1901:  ``{\it Nahm equations, $N=1^\star$ domain walls, and D-strings in $AdS_5\times S^5$},''

1902:   JHEP {\bf 0107}, 041 (2001)

1903:   [arXiv:hep-th/0007067].

1904:   %%CITATION = HEP-TH 0007067;%%

1905: 

1906: \bibitem{kinky} N.~D.~Lambert and D.~Tong,

1907:   ``{\it Kinky D-strings},''

1908:   Nucl.\ Phys.\ B {\bf 569}, 606 (2000)

1909:   [arXiv:hep-th/9907098].

1910:   %%CITATION = HEP-TH 9907098;%%

1911: 

1912: 

1913: \bibitem{ohta} M.~Eto, Y.~Isozumi, M.~Nitta, K.~Ohashi, K.~Ohta and N.~Sakai,

1914:   ``{\it D-brane construction for non-Abelian walls},''

1915:   arXiv:hep-th/0412024.

1916:   %%CITATION = HEP-TH 0412024;%%

1917: 

1918: \bibitem{witten} E.~Witten,

1919:   ``{\it Solutions of four-dimensional field theories via M-theory},''

1920:   Nucl.\ Phys.\ B {\bf 500}, 3 (1997)

1921:   [arXiv:hep-th/9703166].

1922:   %%CITATION = HEP-TH 9703166;%%

1923: 

1924: \bibitem{hh}  A.~Hanany and K.~Hori,

1925:  ``{\it Branes and N = 2 theories in two dimensions}''

1926:   Nucl.\ Phys.\ B {\bf 513}, 119 (1998)

1927:   [arXiv:hep-th/9707192].

1928:   %%CITATION = HEP-TH 9707192;%%

1929: 

1930: \end{thebibliography}

1931: 

1932: 

1933: \end{document}

1934: